UC Riverside

Drug discovery is a costly and time-consuming process. It has four major stages: early drug discovery, pre-clinical phase, clinical trials, and regulatory approval. Early drug discovery is the steps of target validation, hit identification (hit ID), hit-to-lead, and lead optimization for new chemical entities (NCEs) and therapeutic targets. Basic research identifies and validates disease-relevant targets. Hit ID is the process of testing a group of compounds against a target protein. Thermal shift assay (TSA) can detect slight changes in protein melting point and stability, and, therefore, be used to screen for hit ID. X-ray crystallography studies can determine the binding between target protein and hits. In vitro enzymatic assays can confirm the inhibition of hits and measure IC50 (inhibitory concentration 50%). Hit-to-lead can be done by ‘SAR by catalog’, in which the features of the hits are used to identify commercially available compounds. During the hit-to-lead stage, hits are developed to improve the binding affinity to move along as a final drug.Structural-based drug discovery (SBDD) is one of the early-stage drug discovery approaches and has been wildly used in academic and industrial settings since the 1990s. SBDD uses small molecules that follow Lipinski's Rule of Five that ensures that the small molecules can be administered orally once they become drugs. Fragment-based lead discovery (FBLD) evolved from SBDD. Fragments follow the Rule of Three that helps reduce off-target effects and for a broad search of chemical diversity using smaller libraries in the low thousands of compounds. In silico artificial intelligence-computer-aided drug discovery (AI-CADD) is used to identify hits and develop leads using computer tools and the simulation of human intelligence processed by machines. Post-translational modifications (PTMs) are the last stage in protein biosynthesis, being either reversible or irreversible chemical events occurring after protein translation. One type of PTM is ubiquitin-like protein is the Small-Ubiquitin-Related Modifier (SUMO) modification. SUMOylation of a target protein occurs via an enzyme cascade that involves three steps: First, the SUMO E1 activation enzyme, the Aos1/Uba2 complex, activates matured SUMO protein and form a high-energy thioester bond to the Uba2 subunit. Next, the activated SUMO is transferred to cysteine 93 of Ubc9 (SUMO E2 conjugating enzymes). Ubc9 conjugates SUMO onto the substrates, or SUMO E3 enzyme, such as Pias1, often aids in this conjugation event to increase substrate specificity. SUMOylation has been studied extensively and found involved in cancer as “non-oncogene addiction”. SUMO enzymes are not mutated or classical oncogenes, but cancer cells essentially depend on them for survival. Notably, non-oncogene addiction may provide novel targets for intractable and undruggable cancers. Another target for cancer is the dual-specificity tyrosine phosphorylation-regulated kinase 1A (Dyrk1a). It plays multiple roles in animal development, with critical effects on neuronal cell cycle regulation and cancer development. We hypothesis that targeting the SUMOylation modification and Dyrk1a will impact cancer cells as therapeutic treatments. Utilizing SBDD, FBDD, and AI-CADD, we have identified a small molecule, AW2 F5, targets the ATP binding pocket and inhibits Aos1/Uba2 with an IC50 of 497.7 nM. We have identified a fragment, LC B3, covalently bind to the backside of Ubc9 as a potential degrader. We also have found a small molecule, DTP C1, targets the ATP binding pocket and inhibit Dyrk1a with IC50 of 35.7 nM. In the future, each compound will be optimized and tested in cancer cells for their therapeutic effects.