Design and synthesis of small molecules as inhibitors of YTHDF proteins

In this work, I designed and synthesized two different classes of compounds capable of binding to the YTH domain of the YTHDF1-3 proteins. The first class of inhibitors includes ebselen analogs, an organoselenium compound that emerged as the top hit from a high-throughput screening of an in-house compound library; the second class is derived from a computational study. The YTHDF1-3 reader proteins are a class of three highly conserved paralogs that selectively recognize and bind the m6A modification on mRNA. They represent an emerging target in cancer research because m6A regulates several critical biological functions, triggering a cascade of events or directly modulating almost every aspect of tumorigenesis and cancer progression. The interest in targeting these proteins is a growing and recent cutting-edge focus. At the beginning of this thesis, no inhibitors capable of interfering with the m6A recognition process of the YTHDF1-3 proteins were described. Ebselen is a covalent drug capable of forming a selenium-sulfur bond with the cysteine residue located near the YTH domain of the YTHDF1-3 proteins, thereby interfering with the m6A binding process. This work investigated the role of the covalent warhead, which can be either a selenium or a sulfur atom. Furthermore, supported by covalent docking calculations, several analogs were designed and synthesized by modifying the molecular structure of ebselen to improve potency, selectivity, and physicochemical properties. The analogs were purified, fully characterized (NMR spectroscopy, MS spectrometry) and were then tested in an homogeneous time-resolved fluorescence (HTRF) assay to assess their affinity for the target, resulting in low single digit micromolar IC50. In addition, selected compounds were co-crystallized with the YTH domain of the YTHDF1 protein to further demonstrate the covalent binding mechanism of action and to study the effects of the binding on protein conformation. The second hit compound studied in this thesis (m6A71) is able to non-covalently bind the YTH domain of the YTHDF1 protein. The molecular structure of this hit was analyzed to determine the pharmacophore and several analogs were synthesized. In addition, docking calculations were performed to aid in the design of selected analogs and to initiate the lead optimization. The obtained analogs were then evaluated in an HTRF assay and other orthogonal biochemical assays (e.g. RNA electrophoretic mobility shift assay - REMSA), providing valuable insight into the molecular structural features required for the target interactions, that could be further explored and optimized.