Primary liver cancer (PLC) displays high morbidity and mortality in patients with this disease. The latest cancer statistics report shows that 905,677 new cases are diagnosed, and 830,180 patients die from this disease, every year. These figures place PLC as the sixth highest incidence and third highest mortality rate among all cancers [1]. The 3-year survival rate of patients diagnosed with primary liver cancer is approximately 50% [2]. Hepatocellular carcinoma (HCC) is the major form of PLC observed clinically. In early stages, curative treatments for HCC, such as tumor resection, ablation, and liver transplantation can be used for treatment of this disease [3]. However, a large proportion of HCC patients initially present with advanced disease and thus have limited therapeutic options. Systemic drug therapy is one of the approaches to treat advanced HCC, but first-line treatments for advanced HCC, such as sorafenib or lenvatinib, only extend survival by approximately 3 months [4]. Despite tremendous advances in treatment in recent years, the overall prognosis of liver cancer remains poor. Therefore, it is crucial to explore novel biomarkers with prognostic value for prognosis HCC patients.

Ubiquitination is an important post-translational modification that involves the coordinated activity of ubiquitin activating enzymes (termed E1), ubiquitin conjugating enzymes (E2), and ubiquitin ligases (E3). These molecules are responsible for covalently conjugating single or multiple ubiquitin molecules to target proteins [5]. Ubiquitination alters important properties of target proteins, including target protein activity, cellular localization, half-life, receptor endocytosis, and protein-protein interactions [5, 6, 7]. Similar to other post-translational modifications, ubiquitination is a reversible process that is dynamically regulated by ubiquitinases and deubiquitinases (DUBs). Thus, dysregulation of the ubiquitination process can lead to impairment of many important cellular processes and contribute to serious diseases, including cancer [8].

DUBs cleave ubiquitin chains from protein substrates, and owing to structural differences within their catalytic domains, DUBs can be divided into 5 families: Ubiquitin C-Terminal Hydrolases (UCHs), Ubiquitin-Specific Proteases (USPs), Machado-Joseph Disease Protein Domain Proteases (MJDs), Ovarian Tumor Proteases (OTUs), and JAMM Motif Proteases [9]. Among these, USPs is the largest family and has the most diverse structures. Moreover, USPs also belong to the cysteine protease family [9]. Growing evidence suggests that DUBs act as important regulators of immune response, making the development of new drugs targeting USPs a promising therapeutic avenue for the treatment of cancer [10]. Loosdregt et al. [11] found that USP7 can control Treg cell number and function by regulating Foxp3 expression in this type of T lymphocyte. Liu et al. [12] found that USP18 plays a key role in T cell activation and Th17 generation, and further affects NF-$\kappa$B and NFAT activation. Zou et al. [13] observed that inhibition of USP15 induces tumor cell apoptosis and promotes anti-tumor T cell activation. These findings suggest that DUBs, especially USPs, can be considered promising targets for cancer immunotherapy. However, there is currently a lack of comprehensive studies on USPs in HCC, especially in their potential impact on immunotherapy. To our knowledge, this study is the first to explore the expression and prognostic value of 39 USPs members in liver cancer. Furthermore, we also focused on the immunological role of a representative USP member, specifically, USP36.

Ubiquitination is one of the most common and important post-translational protein modifications [27, 28], and the ubiquitin-proteasome system is involved in the degradation of more than 80% of proteins in cells [29]. In addition, ubiquitination is also involved in diverse processes such as regulation of apoptosis, cell cycle progression, endocytosis, and transcription [5, 27]. Ubiquitination and deubiquitination have been observed to be dysregulated in cancer and these processes play multiple roles in cancer-related pathways [30]. Interestingly, ubiquitination is a dynamically reversible process. DUB can effectively downregulate the ubiquitination of key target proteins, and thus further serve to control cell signaling events and cell fate.

At present, a large number of studies have shown that dysregulation of USPs are common molecular events in a variety of cancer types [31] and several studies have sought to clarify the role of USPs in liver cancer. For example, USP7 can stabilize the E3 enzyme TRIP12 and form a complex that promotes HCC cell growth and is significantly associated with HCC malignant phenotype [32]. USP22 promotes hypoxia-induced HCC stemness by regulating HIF-1$\alpha$ [33]. However, we are unaware of a systematic study focused on the role of USPs family members in hepatocellular carcinoma and that this prompted us to seek to better understand an underlying mechanism(s) of USPs in HCC pathobiology.

Our analysis showed that the vast majority (36/39) of USPs were differentially expressed in HCC, suggesting that aberrant protein ubiquitination is prevalent in HCC. We further analyzed the prognostic value of these abnormally expressed USPs in HCC and found that more than ten USPs were linked to HCC prognosis. Importantly, this analysis indicated that USP36 and USP39 can simultaneously, and independently, affect OS, DFS, DSS and PFS in HCC patients, and support our conclusion that increased expression of these two genes are poor prognostic factors in HCC.

Previous studies explored a potential USP39 mechanism of action in HCC. For example, USP39 and TRIM26 co-regulate ZEB1 ubiquitination which controls HCC progression [19]. USP39 promotes HCC progression by deubiquitinating SP1 protein and stabilizing SP1 [21]. In addition, high expression of USP39 was significantly associated with age, tumor status, advanced pathological stage, T stage and higher histological grade in HCC [34].

Previous studies have shown that USP36 can participate in the progression of a variety of cancer types [22, 23], and USP36 can stabilize the protein stability of the hepatoma promoting protooncogene c-myc through dysregulation of its ubiquitination [24, 25]. However, USP36 has not been studied in HCC. This lead us to explore the clinical significance and potential mechanism of USP36 in HCC. We found that in HCC, USP36 was highly expressed in three GEO databases (GSE14520, GSE36376, GSE64041) and validated this finding in 10 matched HCC tumor/normal tissue pairs. These studies revealed that USP36 mutations are uncommon in HCC samples, but that abnormal gene methylation may be one of the reasons for altered USP36 expression in this tumor type.

We also explored the correlation between USP36 and 6 clinicopathological features. Results obtained showed no significant difference in USP36 expression linked to different genders, ages, or lymph node metastasis. However, a significant correlation was noted with tumor stage and tumor grade as increased USP36 expression is more common in higher tumor stage and grade in HCC. This finding prompts us to speculate that USP36 may be involved in HCC progression and, more importantly, expression of USP36 was significantly increased in HCC tumors with TP53 mutation. Previous studies have shown that TP53 is the most frequently mutated gene in HCC and is associated with poor prognosis [35, 36, 37]. Moreover, previous studies have explored a potential interaction between USPs and TP53 gene mutations. For example, the USP22/HIF1$\alpha$ positive feedback loop promotes glycolysis and stemness after TP53 mutation in HCC [33], and USP10 promotes cancer cell proliferation in the presence of TP53 mutations [38]. Taken together, these findings support the notion that USP36 may synergize with TP53 mutations to promote HCC progression. However, whether USP36 plays an important role in TP53-mutated HCC is a scientific question worthy of attention in future studies.

We also analyzed the potential signaling pathways that are affected by USP36 in HCC by using GSEA. Results indicated that the USP36 high-expression group there was significantl enrichment in KEGG UBIQUITIN MEDIATED PROTEOLYSIS, KEGG ENDOCYTOSIS, and KEGG CELL CYCLE and other ubiquitination-related and cancer-related pathways [39, 40]. Studies have shown that ubiquitination is closely related to the cell cycle and endocytosis and that this dysregulation can contribute to cancer progression [41, 42, 43]. Therefore, we speculate that USP36 may affect HCC progression by affecting certain key cyclin expression or receptor endocytosis by dysregulating deubiquitination.

Tumor cells and immune cells, as well as many other cellular and acellular components, make up the complex tumor microenvironment (TME). The TME is highly heterogeneous and dynamic, and can ultimately impact the course of tumor progression [44]. Many studies have shown that TME plays a significant role in HCC and is a very promising potential therapeutic target [45, 46, 47, 48, 49]. Previous studies have shown that ubiquitination plays a crucial role in various physiological processes, including innate and adaptive immunity [8]; thus, the crosstalk between ubiquitination and TME deserves further exploration. It has been reported that USP36 plays an important role in microbial immunity [50] leading us to explore a potential correlation between USP36 and the TME. We used TIMER1.0 to explore the correlation between USP36 expression and six immune cells found in the HCC TME. Notably, USP36 expression showed the strongest correlation with macrophages in the TME. Previous studies have shown that hepatic macrophages contribute to the formation of HCC by suppressing antitumor immunity [51], and that high macrophage infiltration is generally linked to poor HCC prognosis [52]. The potential role of macrophages in HCC is consistent with our analysis of the possible tumor promoting effect of USP36 in this tumor type, suggesting that USP36 may promote HCC by affecting TME components. In addition, our results suggest that USP36 is positively correlated with 16 immune checkpoint molecules, including PDCD1 and CTLA4. Moreover, the IC50 of chemotherapy drugs such as sorafenib, rapamycin, docetaxel and methotrexate were significantly increased in the USP36 high expression group. In conclusion, these results provide some theoretical basis for the use of USP36 expression to help guide therapeutic regimen choices in HCC patients.

However, we must acknowledge some limitations in the current study. First, although we used the authoritative TCGA database, some members of the USPs were not included in the discussion of this study. Second, most of the results of this study were obtained by bioinformatics methods, and there were no in vitro and in vivo experiments and clinical data to verify. In the future, we will focus on the role of USPs members not included in the discussion in HCC. Third, our analyses are all based on the TCGA database, which may reduce the reliability of our findings.

In conclusion, we comprehensively analyzed the expression and prognostic value of USPs in HCC, and analyzed the clinicopathological features of USP36 as well as the potential mechanism of action in detail. We found that USP36 expression is closely related to the tumor microenvironment, which has potential value in clinical application. Future studies should focus on the oncogenic mechanism of USP36 in HCC as a means to independently validate our findings regarding USP36 in this cancer type.

This study is the first to comprehensively analyze the gene characteristics and prognostic value of USPs members in HCC, and focus on the potential mechanism and immunological role of USP36 in HCC. These findings provide new horizons for future research.

ESTIMATE, Estimation of STromal and Immune cells in MAlignant Tumor tissues using Expression data; qRT-PCR, Real-Time Quantitative Reverse Transcription PCR.

YG and LL contributed to the study design. YG and WS were responsible for the writing and proofreading of the article. JS is responsible for data analysis and download. JL and KH took care of the experimental part and graphic editing. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

Ethical approval was obtained from the Ethics Committee of the First Affiliated Hospital of Anhui Medical University (No.Quick-PJ2020-16-23).

Not applicable.

This research received no external funding.

The authors declare no conflict of interest.