Targeting ISG15 and USP18

Division Of Basic Sciences - Nci

Investigators

Linked publications & trials

Abstract

USP18 is the only ISG15-specific protease identified to date although several DUBs have been reported to be cross-reactivity with ISG15 in vitro, such as USP2, USP14 and USP21. For development of a selective USP18 activity-based probe (ABP), we have identified residues in ISG15 that can be mutated to enhance binding and selectivity towards USP18 over other ISG15-reactive deubiqutinases (DUBs) through positional scanning analysis using both a combinatorial peptide library and in silico modeling. For example, our initial results showed that R153Agb and H90F mutations might enhance binding affinity and selectivity towards USP18 over other DUBs. We are currently generating these ISG15 mutants using a combination of solid-phase peptide chemistry and native chemical ligation strategies. After successful preparation of ISG15-based probes, we will perform activity-based protein profiling (ABPP) of USP18 in different cancer cell lines and tumor samples of diverse origin. This study will inform us about the tumor types and states in which USP18 is highly activated and provide insights into the cellular context that USP18 inhibitor can have translational potential. We are also developing selective small molecule inhibitors of USP18. We have developed and optimized a biochemical assay that is suitable for high-throughput screening to identify catalytic inhibitors of USP18. We are currently screening NCI's vast natural product library that encompasses structurally and functionally diverse chemical entities. In parallel, we are conducting virtual ligand screening for non-covalent inhibitors. Two available crystal structures of mUSP18 and mUSP18-ISG15 complex were used to build a single 4D flexible receptor and virtually screen 160,000 commercially available compounds from the ChemBridge DIVERSet library using Molsoft ICM software. After filtering by two docking scores, compounds were clustered by chemotype and predicted binding poses of individual hit compounds were visually inspected. We identified several groups of substituted phthalimides and other chemotypes that were predicted to make hydrogen bonds to the same residues of the protein within the active site. A total of 47 compounds were selected on the basis of diverse chemical structure and high docking score, and were purchased for further evaluations. These compounds are currently under evaluation, though initial results showed two compounds so far with inhibitory activity in our biochemical kinetic assay at a concentration of 30 uM. We are following up with these two lead molecules to measure binding affinity and dose-dependent inhibition as well as ruling out false-positive mechanisms such as aggregation. To evaluate the potential covalent targeting of USP18, we have developed an intact-protein mass spectrometry (MS) assay. A shift in protein mass indicates corresponding covalent labelling of the protein and an emphasis on binding stoichiometry is used to triage compounds. We have obtained commercially available libraries of covalent fragments: vinyl sulfones (200), acrylamides (2240), and a-chloroacetamides (1920). As a pilot study, USP18 was incubated with 5 fragments per pool and the resulting mixtures were subjected to LC-MS analysis. From the screening with vinyl sulfone-containing fragments, we found several covalent fragments that modify USP18. To avoid hyperreactive and promiscuous compounds, we will combine our USP18 MS screening with high-throughput thiol-reactivity assay and counter-screen against other cysteine proteases such as cathepsins. Selective covalent hits from the initial screen will be further evaluated by LC-MS/MS to identify the site of modification and by in vitro fluorogenic assay to determine the inhibitory activity. To identify small molecule binders to USP18, we are also employing small-molecule microarray (SMM) screening. Out of 24k compounds we screened so far, 97 compounds were identified as hits for a hit rate of 0.4%. We purchased these 97 compounds and are currently validating them using our biochemical assay and additional biophysical assays such as surface plasmon resonance (SPR) and NMR waterLOGSY to rule out false-positives and aggregators. Once we have identified lead inhibitors, we will drive the mechanistic, structural, and biological evaluation of USP18 inhibitors through a concerted effort of independent and collaborative research. Specifically, my laboratory will carry out chemical optimization, biochemical, biophysical and cell biological analyses of these molecules. Lead molecules showing activity will be prioritized based on high potency, selectivity, and synthetic tractability. Active molecules will be validated in functional assays that probe protein ISGylation and IFN pathways. Once validated, probes will be used to define USP18's role in cancer development and its therapeutic potential.