Primary liver cancer ranks among the most prevalent cancer types and stands as one of the leading causes of cancer-related fatalities worldwide [1]. Hepatocellular carcinoma (HCC) emerges as the predominant form of primary liver cancer, exhibiting a notably high occurrence in Africa and Asia [1]. Despite its widespread prevalence, effectively treating HCC remains an immense challenge, resulting in a dismal long-term prognosis, especially for patients in advanced disease stages [2]. While radical treatment modalities like surgical tumor excision or liver transplantation are reserved for early-stage cases, therapeutic options for advanced HCC remain limited [2]. Presently, sorafenib stands as the sole approved systemic chemotherapy for HCC, albeit with modest efficacy [3]. Conventional chemotherapeutic agents such as cisplatin or 5-fluorouracil yield unsatisfactory outcomes in the majority of patients [3], partly due to the genomic instability and intricate morphology of HCC cells, which render them inherently resistant to standard chemotherapies or prompt them to develop resistance shortly after an initial response [3]. Given its intrinsic chemoresistance and acquired drug resistance, there exists an urgent imperative to uncover novel compounds capable of sensitizing HCC to conventional therapeutic agents [4]. In recent years, it has been found that the level of ubiquitination of some carcinogenic protein in HCC cells has been significantly reduced [4,5]. The regulation of deubiquitinating enzymes (DUBs) in the ubiquitin system is very critical [4]. The up-regulation of the expression and function of DUBs leads to the abnormally active deubiquitination of some oncogenic proteins, which in turn resists the degradation of the ubiquitin system. Therefore, targeting deubiquitination-modifying enzymes to relevant oncogenic proteins may become a novel therapeutic strategy to overcome HCC [6].

Ubiquitin-specific protease 10 (USP10) stands as a member of the largest subgroup of DUBs referred to as ubiquitin-specific proteases (USPs). Its multifaceted role encompasses counteracting ubiquitin ligase activity, modifying ubiquitin chains, and salvaging ubiquitin from proteins [7]. Extensively investigated, USP10 exerts its influence over a gamut of cellular mechanisms, including DNA repair, cell-cycle orchestration, autophagy, as well as immune and inflammatory responses [[8], [9], [10], [11]]. Notably, USP10's overexpression has been closely linked to the promotion of proliferation and metastasis in diverse malignancies, among them HCC and colon cancer, establishing its potential as a viable target for therapeutic intervention within the oncological realm [12,13].

Nonetheless, despite the ongoing efforts, the landscape concerning USP10 inhibitors still stands quite limited. Notably, PR-619 (1) with 3,5-dithiocyanatopyridine scaffold offers reversible inhibition of USP10, but its role as a broad-spectrum DUB catalytic inhibitor diminishes its specificity (Fig. 1) [14,15]. Quinazolin-4-amine derivative spautin-1 (2, IC50 = 0.6 μM) stands as a competitive inhibitor of USP10, identified through phenotypic-based screening (Fig. 1). However, it also demonstrates cross-reactivity against ubiquitin-specific proteases 13 (USP13), while the direct binding of spautin-1 to USP10 has yet to be unequivocally established [16,17]. The 1,2,3,4-tetrahydroacridine derivative HBX19818 and 3-nitrothiophene derivative P22077 were highlighted for their direct binding to and inhibition of USP10 (Fig. 1). Their capacity to impede the growth of acute myeloid leukemia bearing fms related receptor tyrosine kinase 3 (FLT3) mutation is noteworthy. However, their dual effect on ubiquitin-specific proteases 7 (USP7) catalytic activity, with IC50 values of 57 μM and 10 μM respectively, warrants attention [18,19]. Wu-5 was another reported USP10 inhibitor with 3-nitrothiophene scaffold which was discovered from an in-house compound library screening (Fig. 1). Its potential to surmount FLT3-inhibitor resistance and amplify the anti-acute myeloid leukemia (AML) impact of crenolanib is intriguing [20]. Nevertheless, structural resemblances between P22077 and Wu-5 fuel concerns about their USP7 selectivity. In summation, the existing cohort of reported USP10 inhibitors, primarily sourced from phenotypic screening, exhibits suboptimal selectivity. Considering that protein deubiquitination is regulated by multiple DUBs and non-selective inhibitors cannot be used as optimal tools, there is an urgent need for the development of USP10 inhibitors with strong potency and high selectivity for the study of USP10-related regulatory mechanisms and its role in HCC.

In our earlier research, we uncovered a noteworthy association between increased USP10 levels and elevated levels of Yes-associated protein (YAP) and transcriptional co-activator with PDZ-binding motifs (TAZ). Furthermore, our findings indicate that USP10 plays a pivotal role in driving HCC progression by stabilizing YAP/YAZ through deubiquitination [13]. Nonetheless, the deficiency of selective USP10 inhibitors has significantly constrained the potential intervention strategies. In this work, we constructed a workflow for the discovery of USP10 inhibitors. Briefly, we predicted the structure of USP10 catalytic domain through artificial intelligence (AI)-driven protein structure prediction method based on the amino acid sequence from Uniport database. Next, we used mixed-solvent molecular dynamics to further explore the targetable pocket of USP10, and subsequently performed a virtual screen to discovery USP10 inhibitors with novel scaffold to target this pocket. After the structure optimization, we identified a compound labeled as D1, exhibiting USP10 inhibition efficacy within low micromolar concentrations. Our investigations demonstrated that this newly discovered ligand possesses the capability to directly bind to USP10 and effectively inhibit its deubiquitination activity, and thereby affect biological processes. It's noteworthy that our study signifies the pioneering achievement of identifying a novel USP10 inhibitor through the prism of structure-based drug design, underscoring its potential as a chemical tool for interrogating the intricate function of USP10.