Transcriptome data (RNA-seq) and corresponding clinical information of the Liver Hepatocellular Carcinoma (LIHC) project were downloaded from The Cancer Genome Atlas (TCGA) portal (https://portal.gdc.cancer.gov).

The initial dataset comprised 424 clinical samples, including 371 tumor tissues and 53 paired adjacent normal tissues. Samples with incomplete clinical information were excluded from subsequent survival and clinical correlation analyses. To augment the normal liver tissue sample size, transcriptome data from 110 normal liver samples were obtained from the Genotype-Tissue Expression (GTEx) project database (https://commonfund.nih.gov/GTEx). All gene expression data were normalized to Transcripts Per Million (TPM) format. For downstream analyses, the TPM values were transformed using log2(TPM + 1) to approximate a normal distribution. All data processing was performed using R software (version 4.2.1; https://www.R-project.org).

The Xiantao Toolbox (https://xiantaozi.com/) platform provides a non-coding suite of tools for multi-dimensional bioinformatics analysis without requiring programming expertise. It was utilized in the present study for initial data exploration and specific analytical modules.

TISCH database. Tumor Immune Single-cell Hub (TISCH, http://tisch.comp-genomics.org/) is a curated database dedicated to the tumor microenvironment (TME), providing single-cell RNA sequencing (scRNA-seq) data across various cancer types. It was consulted to explore the cellular composition of the TME in HCC at a single-cell resolution.

BioGRID database. Biological General Repository for Interaction Datasets (BioGRID, https://thebiogrid.org) is a publicly accessible repository of protein, genetic and chemical interactions. This database provides over 2.8 million protein types and genetic interactions, and more than 30.000 chemical interactions, and this database was queried to identify known and predicted protein-protein interactions (PPIs) involving PSMD6.

UALCAN database. UALCAN is a comprehensive web-based bioinformatics toolbox platform (https://ualcan.path.uab.edu/) for analyzing cancer OMICS data, particularly from TCGA. It enables in-depth gene expression analysis, patient survival analysis, and promoter methylation profiling across different tumor subgroups. This database was used to validate PSMD6 expression patterns and assess its correlation with key clinical parameters (for example, cancer stage and tumor grade).

SANGERBOX database. SANGERBOX is an integrated bioinformatics analysis platform (http://vip.sangerbox.com) offering a user-friendly interface for various analyses, including differential expression, pathway enrichment and Weighted Gene Co-expression Network Analysis (WGCNA). It was employed for specific functional enrichment analyses.

Kaplan-Meier Plotter. Kaplan-Meier Plotter data repository (https://kmplot.com/analysis/) is designed to assess the effect of gene expression on survival outcomes across multiple cancer types using data from sources such as GEO and TCGA. It was used specifically to evaluate the prognostic significance of PSMD6 expression in terms of overall survival (OS) and disease-specific survival (DSS) in the TCGA-LIHC cohort. The platform employs a Cox proportional hazards model, and hazard ratios with 95% confidence intervals and log-rank P-values are reported.

STRING database. STRING is a database of known and predicted PPIs (https://string-db.org). It was used to construct a PPI network centered on PSMD6, with a confidence score threshold set to ensure high-quality interactions.

TIMER database. TIMER (https://cistrome.shinyapps.io/timer/) is a web server for comprehensive analysis of immune cell infiltrates across diverse cancer types using TCGA data. It was utilized to investigate the correlations between PSMD6 expression levels and the abundance of six immune cell populations (B cells, CD4⁺ T cells, CD8⁺ T cells, neutrophils, macrophages and dendritic cells) within the HCC TME, with adjustment for tumor purity.

The human immortalized hepatic epithelial cell line (THLE-2) and three human liver cancer cell lines were procured from Zhejiang Meisen Cell Technology Co., Ltd. (THLE-2, cat. no. CTCC-004-0030; HepG2, cat. no. CTCC-DZ-0057; MHCC97-L, cat. no. CTCC-400-0194; MHCC97-H, cat. no. CTCC-0395-Luc1). The 4 cell lines was authenticated using short tandem repeat profiling. The obtained STR profile was compared with reference databases (DSMZ) to confirm identity. Cells were routinely tested for mycoplasma contamination and used within 20 passages after thawing. All cell lines were cultured in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.), supplemented with 10% heat-inactivated fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin-streptomycin (Gibco; Thermo Fisher Scientific, Inc.). Cells were maintained in a humidified incubator at 37˚C with 5% CO₂. Regular mycoplasma testing was performed to ensure cell line authenticity and absence of contamination.

Total protein was extracted from cultured cells using RIPA lysis buffer (Guangzhou Biolight Biotechnology Co., Ltd.) containing protease and phosphatase inhibitors (Tiandz, Inc.). Protein concentration was determined using a bicinchoninic acid (BCA) assay kit. Equal amounts of protein (typically 20-30 µg per lane) were separated by 10% SDS-polyacrylamide gel electrophoresis and subsequently transferred onto polyvinylidene difluoride membranes (Seebio; https://www.seebio.cn/). After blocking with 5% (w/v) bovine serum albumin (cat. no. BS114; Biosharp Life Sciences) in Tris-buffered saline with 0.1% Tween-20 (TBST) for 1 h at room temperature, the membranes were incubated overnight at 4˚C with the primary antibody against PSMD6 (1:1,000; cat. no. Ag3251; Proteintech Group, Inc.) and GAPDH as a loading control (1:50,000; cat. no. 60004-1-Ig; Proteintech Group, Inc.). Following three washes with TBST, the membranes were incubated with an appropriate horseradish peroxidase (HRP)-conjugated secondary antibody (1:50,000; cat. no. BL003A; Biosharp Life Sciences) for 1 h at room temperature. Protein bands were visualized using an enhanced chemiluminescence (ECL) detection kit (MilliporeSigma) and imaged with a chemiluminescence imaging system. Densitometric analysis of the bands was performed using ImageJ software (version 1.8.0.; National Institutes of Health). All experiments were independently repeated at least three times.

Total RNA was extracted from cultured cells using TriQuick Reagent (Beijing Solarbio Science & Technology Co., Ltd.) according to the manufacturer's instructions. RNA concentration and purity (A260/A280 ≥1.8) were measured using a NanoDrop spectrophotometer (Thermo Fisher Scientific, Inc.). Complementary DNA (cDNA) was synthesized from 1 µg of total RNA using a 5X RT SuperMix for qPCR kit (Vazyme Biotech Co., Ltd.). qPCR was performed using a 2X SYBR Green qPCR Master Mix (Vazyme Biotech Co., Ltd.) on an iQ5 Multicolor Real-Time PCR Detection System (Bio-Rad Laboratories, Inc.). The PCR cycling conditions were as follows: Initial denaturation at 95˚C for 30 sec, followed by 40 cycles of 95˚C for 5 sec and 60˚C for 30 sec. The relative mRNA expression level of PSMD6 was calculated using the 2^-ΔΔCq method (20), with GAPDH as the internal reference gene. The specific primer sequences used are listed in Table I. Each experiment was performed in triplicate wells and repeated independently three times.

Bioinformatics data analysis was primarily conducted using R software (version 4.2.1 or 3.6.4 as specified for specific packages). Differential expression of PSMD6 between HCC tumors and normal tissues was assessed using the Wilcoxon rank-sum test. Correlations between PSMD6 expression and immune cell infiltration levels were evaluated using Pearson's or Spearman's correlation coefficients via the corr.test function in the Psych R package (v2.1.6; CRAN.R-project.org/package=psych). Survival analysis was performed using the Kaplan-Meier method, and differences in survival curves were compared using the log-rank test via the Survival R package. For in vitro experiments, data are presented as the mean ± standard deviation (x̄±s) from at least three independent experiments. Statistical comparisons between two groups were analyzed using independent two-tailed unpaired Student's t-test. Comparisons among multiple groups were performed using one-way analysis of variance (ANOVA) followed by Dunnett's multiple comparison test. P<0.05 was considered to indicate a statistically significant difference.

Investigation was initiated by assessing the pan-cancer expression profile of PSMD6 to determine its potential relevance in oncogenesis. To tackle this, the pan-cancer expression of PSMD6 was assessed utilizing SANGERBOX database, and it was revealed that PSMD6 was significantly dysregulated across multiple cancer types, being upregulated in 15 and downregulated in 15 other malignancies (Table II, Fig. 1A). Consistent findings were gained through analyzing with UALCAN data repository (Fig. 1B). This pan-cancer aberration highlighted PSMD6 as a gene of general interest in cancer research.

Subsequent focused analysis on LIHC from the TCGA cohort demonstrated a pronounced overexpression of PSMD6 in tumor specimens (n=371) compared with adjacent normal tissues (n=50; P<0.001; Fig. 2A and B).

To evaluate the clinical relevance of PSMD6 overexpression, its correlation with key clinicopathological parameters was examined using Xiantao tools. PSMD6 levels were significantly elevated in HCC tissues across all pathology grades (G1-G4), T stages (T1-T4), nodal involvement (N0/N1), metastasis status (M0/M1), as well as irrespective of patient weight, sex and ethnicity (all P<0.05) (Fig. 3A-G). This consistent upregulation across diverse clinical subgroups underscores the potential of PSMD6 as a robust biomarker associated with HCC progression and aggressiveness.

Given the strong association of PSMD6 with advanced disease stages, the prognostic value of PSMD6 expression in HCC was next evaluated. A 5-year survival analysis was conducted on patients using the UALCAN database, and the results showed that patients with high PSMD6 expression (n=90) had a significantly lower probability of overall survival compared with those with low expression (n=275; P<0.05; Fig. 4A).

In the subgroup analysis stratified by pathology grade, the best prognosis was found in grade 1 patients with a low expression level of PSMD6, and the worst prognosis was found in grade 4 patients highly expressing PSMD6. In the subgroup analysis stratified by ethnicity, African Americans with high PSMD6 expression had the worst prognosis, while Caucasians with low PSMD6 expression had the best prognosis. In the subgroup analysis stratified by sex, female group with high expression of PSMD6 had the most unfavorable prognosis, and female group with low expression of PSMD6 had the best prognosis. All comparisons exhibited statistical significance (Fig. 4B-D).

Further validation using the SANGERBOX database corroborated these findings across multiple survival endpoints. High PSMD6 expression was significantly associated with poorer OS, DSS, disease-free interval and progression-free interval (all P<0.05; Fig. 5).

The robustness of PSMD6 as a prognostic indicator was confirmed by Receiver Operating Characteristic (ROC) curve analysis utilizing the Kaplan-Meier Plotter data repository. Cases were classified into high-expression groups and low-expression groups relying on the mean expression at different quantiles. The analysis revealed that PSMD6 gene expression had a favorable prognostic value (P=0.00039; Fig. 6A), which yielded an area under the curve of 0.877 (95% confidence interval: 0.837-0.916; Fig. 6B), indicating high diagnostic accuracy.

To further validate bioinformatic predictions, in vitro experiments were performed using a normal human hepatic cell line (THLE-2) and three distinct HCC cell lines (HepG2, MHCC97-L and MHCC97-H). RT-qPCR analysis demonstrated that PSMD6 mRNA levels were significantly elevated in all three HCC cell lines compared with the normal hepatic cell line (P<0.0001; Fig. 7).

This transcriptional upregulation was further confirmed at the protein level by western blot analysis, which revealed consistently higher PSMD6 protein expression in the HCC cell lines (Fig. 8). These findings from mRNA and protein levels provide strong evidence for the overexpression of PSMD6 in HCC.

PSMD6 expression correlates with immune cell infiltration in the TME. Tumorigenesis and development usually correlate to infiltration of immune cells including B cell and macrophage. In this context, the TIMER database was utilized to investigate the relationship between PSMD6 expression and the abundance of various immune cell populations within the HCC TME. PSMD6 expression was found to be independent of tumor purity (r=0.007, P=0.89). However, it exhibited significant positive correlations with the infiltration levels of B cells (r=0.191, P=0.00003), CD8⁺ T cells (r=0.1416, P=0.0087), CD4⁺ T cells (r=0.2287, P=1.84x10^-5), neutrophils (r=0.299, P=1.48x10^-8), macrophages (r=0.307, P=6.9x10^-9) and dendritic cells (r=0.268, P=5.45x10^-7) (Table III, Fig. 9A).

To gain deeper, single-cell resolution insights, the TISCH database was used to analyze the enrichment of PSMD6 gene in tumor and its immune microenvironment at the single-cell level. Analysis of the GSE140228_SmartSeq2 and GSE98638 datasets revealed that PSMD6 is predominantly enriched within specific immune cell subsets, including monocytes/macrophages (mono/macro) and natural killer (NK) cells in GSE140228_ smartseq2 single-cell cohort, and in conventional CD4⁺ T cells (cd4tconv) and CD8⁺ T cells in GSE 98638 single-cell cohort (Fig. 9B-D). This pattern suggests that PSMD6 may influence the immune contexture of HCC, potentially contributing to the immunosuppressive microenvironment often observed in advanced tumors.

To elucidate the potential biological functions and pathways through which PSMD6 contributes to hepatocarcinogenesis, comprehensive functional enrichment analysis was performed. Currently, Gene Ontology (GO) functional and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment evaluations are the methods that are used most frequently.

KEGG pathway analysis indicated that PSMD6-coexpressed genes were significantly enriched in crucial cancer-related pathways, including RNA transmit, Ubiquitin mediated proteolysis, Cell cycle, mRNA surveillance pathway, Spliceosome, Herpes simplex virus 1 infection, basal transcription factors, shigellosis, endocytosis and DNA replication (Fig. 10). Concurrently, GO functional analysis highlighted terms such as nucleoplasm, chromosome organization, protein-containing complex and catalytic complex (Fig. 10). The consistent emergence of cell cycle and protein degradation pathways aligns with the known role of the 26S proteasome, of which PSMD6 is a regulatory subunit in controlling cell cycle progression and protein turnover.

Furthermore, a PPI network was constructed to identify potential functional partners of PSMD6. Queries using the BioGRID and STRING databases revealed that PSMD6 interacts with several core proteasomal subunits, including PSMC3, PSMD7, UCHL5, PSMA2 and PSMB7 (P<0.001, Fig. 11).

Moreover, as a PPI network can facilitate an improved understanding of intracellular protein functions and signal delivery, a PSMD6-centered PPI network was constructed using STRING database. Multiple PSMD6-interacting proteins were identified, such as PSMC1, PSMA2, PSMA4 and PSMB7 (Fig. 12), suggesting that its oncogenic effects may be mediated through the regulation of proteasome activity and subsequent impacts on critical cellular processes.

The present multi-faceted analysis demonstrates that PSMD6 is significantly overexpressed in HCC and is strongly associated with malignant progression and poor patient survival. Its influence likely extends to shaping the tumor immune microenvironment. These findings nominate PSMD6 as a promising diagnostic biomarker and a potential therapeutic target worthy of further investigation in HCC.

Primary liver cancer is the seventh-most frequently occurring cancer in the world and the second-most common cause of cancer mortality (21). China has the greatest number of primary liver cases, attributable to both an elevated rate (18.3 per 100,000) and the world's largest population (1.4 billion persons) (22). Specifically, HCC is the dominant type of liver cancer, accounting for ~75% of all liver cancers (23).

With the development of surgical intervention, chemotherapy, targeted treatment and immunotherapy, the survival rate of HCC cases has benefited more from multidisciplinary treatment approaches. However, the survival rate remains low, highlighting a pretty grim situation (24). Therefore, HCC remains a formidable global health challenge, characterized by high mortality and limited therapeutic options for advanced-stage patient.

The quest for reliable biomarkers for early detection and prognosis prediction is therefore paramount. To identify HCC early, current screening and surveillance strategies depend on reliable and convenient serum biomarkers, but only alpha-fetoprotein (AFP) has been extensively applied in clinical settings (25). However, AFP has poor sensitivity and specificity. The present study identified PSMD6 as a potential key player in HCC pathogenesis. Bioinformatics analysis across multiple databases (for example, TCGA and GTEx) consistently revealed significant upregulation of PSMD6 in HCC tissues compared with normal liver tissues. This finding was robustly validated in vitro, where both mRNA and protein levels of PSMD6 were markedly elevated in HCC cell lines (HepG2, MHCC97-L, MHCC97-H) compared with the normal hepatic THLE-2 cell line.

These findings align with and extend previous bioinformatics studies that also identified PSMD6 as a differentially methylated gene with significant prognostic value in HCC (26,27). The consistency of PSMD6 overexpression across diverse cohorts and its strong association with clinical outcomes underscore its potential as a robust diagnostic and prognostic biomarker.

Proteasomes rapidly catalyze a diverse range of biological reactions and then regulate the bioactivity of multiple cells (28). In a previous study, a proteasome system in Caenorhabditis elegans germline was constructed and it was found that PSMD6 downregulation led to reduced proteasome proteolytic activity and cell cycle defect (29). PSMD family members may exhibit an increasing trend in tumors due to the characteristics of extremely high protein metabolic level and very short cell division cycle in tumor cells. The PPI network analysis revealed that PSMD6 interacts with core proteasomal subunits (for example, PSMC3, PSMD7 and PSMA2), underscoring its integral role within the 26S proteasome complex. This interaction network suggests that the oncogenic effects of PSMD6 are likely mediated through the regulation of proteasome activity, impacting a wide array of downstream cellular processes.

Dysregulation of UPS components, including various PSMD family members, is increasingly recognized as a hallmark of cancer, contributing to uncontrolled proliferation and evasion of apoptosis. The functional enrichment analyses (GO and KEGG) of the present study shed light on the potential mechanisms by which PSMD6 contributes to hepatocarcinogenesis. PSMD6 was significantly enriched in essential biological processes such as the cell cycle, ubiquitin-mediated proteolysis and DNA replication.

A pivotal finding of the present study is the significant correlation between PSMD6 expression and immune cell infiltration. PSMD6 levels positively correlated with the abundance of various immune cells, including B cells, CD4⁺ T cells, CD8⁺ T cells, macrophages and dendritic cells. scRNA-seq data from the TISCH database further indicated that PSMD6 is enriched in specific immune subsets including monocytes/macrophages, NK cells and T cells within the HCC microenvironment. This suggests that PSMD6 may influence the functional state of the immune landscape, potentially fostering an immunosuppressive niche that facilitates immune evasion. The role of the immune microenvironment in determining HCC prognosis is well-established, and the current findings position PSMD6 as a novel modulator within this context.

Immunotherapy is important for patients with HCC (30,31). Given the significant connection amid the expression of PSMD6 and immune cell infiltration, patients with HCC with poor prognosis and survival may benefit from immunotherapy. However, whether PSMD6 gene expression affects the effectiveness of immunotherapy warrants further validation.

The present study also noted that Zhou et al (27) previously proposed in their research that PSMD6 and its homolog PSMD11 may participate in liver cancer progression and treatment response by regulating proteasome dependent protein degradation pathways, thereby affecting the stability of key cancer proteins such as p53 and Cyclin. This hypothesis provides an important mechanistic perspective for the findings of the present study. The experimental results further support this viewpoint: PSMD6 expression is significantly upregulated in liver cancer tissues, and its high expression is closely related to poor prognosis in patients. Based on previous studies, it was hypothesized that PSMD6 may promote the development of liver cancer through the following pathways: Enhancing proteasome activity and accelerating the degradation of tumor suppressor proteins; regulating the stability of cell cycle and apoptosis related proteins through the ubiquitin proteasome system; synergistic effect with subunits such as PSMD11 affects the proliferation and invasion ability of liver cancer cells. Future research can further validate the specific regulatory mechanism of PSMD6 in liver cancer by knocking down/overexpressing it and detecting downstream target protein levels, providing new ideas for targeted proteasome subunits in liver cancer therapy.

While the integrated bioinformatics and preliminary experimental data highlight the significance of PSMD6 in HCC, the present study has several limitations. First, the bioinformatic findings, though validated across multiple databases, require further confirmation in larger, prospectively collected clinical cohorts to solidify their clinical translatability. Differences in sample sources, sequencing platforms and batches in public databases may lead to biased results, affecting the comparability of PSMD6 expression. Second, the precise molecular mechanisms by which PSMD6 regulates the cell cycle, immune infiltration and other cancer-related pathways remain to be fully elucidated through detailed functional experiments. Only partial in vitro experimental studies were conducted without tissue validation. No stratified analysis was conducted in the present study based on liver cancer subtypes, staging, or treatment response, which may mask potential heterogeneity and therefore lead to weak clinical correlation analysis Although the co-expression network or pathway enrichment of PSMD6 can be predicted, its upstream regulation (such as methylation, miRNA) and downstream effects in liver cancer still require experimental exploration. Prospective cohorts and animal models are needed to further validate its prognostic value and therapeutic potential.

In conclusion, PSMD6 was found to be significantly overexpressed in HCC and is strongly associated with malignant progression and poor patient survival. Its influence likely extends to shaping the tumor immune microenvironment. These findings nominate PSMD6 as a promising diagnostic biomarker and a potential therapeutic target worthy of further investigation in HCC. Further experiments and clinical samples are needed for verification.