Identification of UPB1 as a tumor suppressor associated with immune infiltration in cholangiocarcinoma

Abstract

BACKGROUND

Cholangiocarcinoma (CCA) remains one of the most aggressive malignancies within the digestive tract, and identifying novel biomarkers and therapeutic targets is essential to improve patient survival. Recent studies have highlighted UPB1 as a predictive biomarker across multiple tumor types, but its biological involvement in CCA development and progression remains unclear.

AIM

To clarify the expression and prognostic value of UPB1 in CCA, its biological functions in tumor growth and immune infiltration.

METHODS

A combination of bioinformatics analyses and clinical data evaluation was applied to investigate UPB1 expression in CCA. Functional assays in cell models and xenograft mouse experiments were carried out to assess the influence of UPB1 on CCA growth. Co-culture assays were utilized to examine how UPB1 expression influences tumor immune infiltration.

RESULTS

Reduced UPB1 expression in CCA was correlated with unfavorable clinical manifestations and poor survival outcomes. In vitro studies demonstrated that UPB1 overexpression enhanced apoptosis and triggered G1/S cell cycle arrest in CCA cells. Reduced expression of BCL-2 and BIRC5, as well as down-regulation of CDK2 and CCNA2, were observed in UPB1 overexpressing cells. In vivo experiments further confirmed that UPB1 exerts a moderate tumor-suppressive effect on CCA xenograft development. Moreover, UPB1 expression showed a positive correlation with tumor-infiltrating lymphocytes, and in vitro co-culture assays suggested that UPB1 overexpression upregulated T-cell activation related transcripts and increased the release of immune factors.

CONCLUSION

UPB1 functions as a tumor suppressor in CCA and is associated with tumor immune infiltration. These results deepen understanding of CCA biology and highlight UPB1 as a promising therapeutic candidate.

Key Words: UPB1; Cholangiocarcinoma; Tumor suppressor; Immune infiltration; Cell cycle; Apoptosis

Core Tip: This study identifies UPB1 as a tumor-suppressor in cholangiocarcinoma (CCA). The expression of UPB1 is downregulated in CCA and correlates with adverse clinical features and worse survival. UPB1 overexpression enhances cell apoptosis and triggers G1/S phase arrest. UPB1 expression shows a positive correlation with tumor-infiltrating lymphocytes, and UPB1 overexpression promotes T-cell activation and immune factor release. UPB1 represents a novel prognostic biomarker and a candidate for further investigation in CCA.

INTRODUCTION

Cholangiocarcinoma (CCA) represents the second most prevalent primary hepatobiliary carcinoma and can be categorized into intrahepatic CCA (iCCA) and extrahepatic CCA (eCCA) based on their anatomical origin. The incidence of CCA is relatively low in Western countries but is highest in Southeast Asia, particularly in Thailand and China[1]. Globally, the incidence of CCA has been steadily increasing, with an average annual rise of 4.4%[2]. Despite advances in diagnosis and treatment, CCA remains one of the most aggressive digestive system tumors, with a 5-year survival of 7%-20%[3]. Surgery is the main curative option for early-stage cases, whereas therapeutic choices for advanced disease remain limited. Although targeted therapies and immunotherapies have been introduced to improve prognosis, long-term survival outcomes remain poor due to the highly aggressive biological behavior of CCA and its propensity for early recurrence[4,5]. Thus, the identification of reliable and specific molecular biomarkers is urgently needed to enable early detection, stratify patient risk, and guide personalized therapeutic strategies.

UPB1, located on chromosome 22q11.2, encodes β-ureidopropionase, a 5-fluorouracil-inactivating enzyme involved in the degradation of pyrimidine bases such as uracil and thymine[6,7]. UPB1 exhibits high enzymatic activity in the liver, and genetic alterations in this gene lead to a rare autosomal recessive disorder, β-ureidopropionase deficiency, characterized by abnormal propionate metabolism[8]. Recent studies have identified UPB1 as a potential prognostic biomarker in several tumor types, including hepatocellular carcinoma (HCC), clear cell renal cell carcinoma, and lung adenocarcinoma (LUAD)[9-11]. Moreover, UPB1 expression has been reported to correlate with cluster of differentiation (CD) 8⁺ T-cell infiltration and to predict the efficacy of immunotherapy in stage III LUAD[12]. Although aberrant UPB1 expression has been linked to prognosis in multiple malignancies, its biological role in CCA remains undefined, and functional studies assessing its effects on tumor cell behavior are still lacking.

In this study, the impact of UPB1 expression on CCA was examined through bioinformatics analyses, cellular and animal models, and investigations of its molecular regulatory mechanisms in cell cycle progression and apoptosis. The association of UPB1 expression with immune infiltration in the tumor microenvironment (TME), particularly tumor-infiltrating lymphocytes (TILs) and immune factors, was further explored to identify potential therapeutic targets and clinical strategies for CCA prevention and treatment. The findings demonstrated that UPB1 functions as a tumor-inhibitory factor in CCA, providing a scientific basis for the development of novel therapeutic approaches and offering more effective treatment options for patients.

MATERIALS AND METHODS

Bioinformatic data acquisition and analysis

The Gene Expression Profiling Interactive Analysis (GEPIA) platform, which integrates transcriptomic information from the Genotype-Tissue Expression project and The Cancer Genome Atlas (TCGA), was utilized to assess UPB1 expression across various cancer types[13]. To further validate UPB1 expression in CCA, four Gene Expression Omnibus (GEO) datasets were analyzed: GSE26566 (104 CCA vs 65 normal bile ducts), GSE32225 (149 CCA vs 6 normal tissues), GSE32958 (30 CCA vs 7 normal tissues), and GSE76311 (91 CCA vs 92 normal tissues). To identify co-regulated genes of UPB1, the cBioPortal repository was queried, applying stringent criteria of Spearman’s correlation coefficient > 0.5 and P < 0.05[14]. Subsequent functional annotation was performed using enrichment analyses based on the Gene Ontology (GO) database and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, with a significance threshold of adjusted P value < 0.05 and a minimum gene count of 2 using the DAVID 6.8 database. For the Gene Set Enrichment Analysis (GSEA), we performed 1000 permutations and considered gene sets with a nominal P value < 0.05 and a false discovery rate q value < 0.25 as significantly enriched. The normalized enrichment score was used to rank the enriched pathways between high- and low-UPB1 expression groups. Finally, gene interaction maps and protein-protein interaction (PPI) networks of UPB1 were generated using the GeneMANIA and STRING platforms to investigate the relationships between UPB1 and its associated genes[15,16].

Tumor immune infiltration analyses

The relationships between UPB1 expression and TILs (including CD4⁺ and CD8⁺ T cells, B lymphocytes, natural killer cells, dendritic cells, macrophages, mast cells, eosinophils, and neutrophils), immunostimulators, immunoinhibitors and chemokines were analyzed using the TISIDB database, an integrated platform for tumor-immune system interactions[17]. Additionally, transcriptomic profiles of CCA samples were obtained from the TCGA database (updated in June 2025) to evaluate the relationships between UPB1 levels, TIL signature genes, and immune checkpoint markers.

Clinical sample collection and patient follow-up

Eighty-four paired tumor and adjacent normal bile duct tissues were collected from patients with pathologically confirmed CCA at Peking Union Medical College Hospital between January 2020 and December 2023. Follow-up was carried out until June 2025, yielding a median monitoring duration of 24.5 months (range, 14.5-38.8 months). Overall survival (OS) was calculated from the date of diagnosis to death or the last recorded visit, while disease-free survival (DFS) was defined as the interval from the initiation of therapy to the first evidence of recurrence or progression. Associations between UPB1 expression and clinicopathological parameters, including sex, age, tumor grade, lymph node status, lymphovascular invasion, serum carbohydrate antigen 19-9 level, and tumor node metastasis (TNM) classification (American Joint Committee on Cancer, 8th edition), were assessed by correlation analysis. Independent prognostic factors for CCA were examined through univariate and multivariate Cox regression analyses. Patients were stratified into high- and low-UPB1 cohorts according to the median H-score. Survival outcomes were visualized using Kaplan-Meier curves to determine the prognostic value of UPB1. The study protocol was approved by the Ethics Committee of Peking Union Medical College Hospital (approval No. I-23PJ274) and conducted in accordance with the Declaration of Helsinki.

Cell culture and lentivirus transfection

RBE, HUCCT1, HCCC-9810, CCLP-1, QBC939, and SNU-869 cell lines were obtained from Yaji Biotechnology (Shanghai, China), with short tandem repeat identification reports provided. All cells were maintained in Roswell Park Memorial Institute (RPMI) 1640 medium enriched with 10% fetal bovine serum (FBS) (Tianhang Biotechnology, China) under standard conditions (37 °C, 50 mL/L carbon dioxide, humidified atmosphere). To generate UPB1-overexpressing and UPB1-knockdown cell lines, Puro-cytomegalovirus (CMV)-UPB1-3 × FLAG (DYKDDDDK) and Puro-CMV-short hairpin RNA (shRNA) targeting UPB1-enhanced green fluorescent protein vectors were introduced into HEK293T packaging cells, and viral particles were harvested three days post-transfection. Target cells were then infected using TUNDUX virus transducers, followed by puromycin selection to establish stable lines. The shRNA sequences employed were: ShRNA1: 5’-UCUACGAUAGCCUUUAUGCGU-3’; 3’-GCAUAAAGGCUAUCGUAGAGG-5’; shRNA2: 5’-AUAAAAGUAGCCAAAGUCCUG-3’; 3’-GGACUUUGGCUACUUUUAUGG-5’. For co-culture experiments, peripheral blood samples were obtained from healthy donors. After dilution, peripheral blood mononuclear cells (PBMCs) were isolated using Ficoll-Paque PLUS (Yuningwei Biotechnology, China). T-cell subsets were purified from PBMCs with the Dynabeads^TM Untouched^TM Human T Cell Kit (Thermo Fisher Scientific, 11344D) and cultured in RPMI-1640 medium containing 10% FBS. In the Transwell system, T lymphocytes were seeded in the upper chamber, while CCA cells were plated in the lower chamber. After five days of co-culture, T cells and supernatants were collected for subsequent analyses.

Cell functional experiments

Cell proliferation was assessed using colony formation and cell counting kit-8 (CCK-8) assays, whereas migratory ability was examined by Transwell migration and wound-healing assays, following previously described protocols[18]. Cell cycle phases and apoptotic fractions were quantified by flow cytometric (FCM) analysis.

Real-time quantitative polymerase chain reaction, western blot and Co-immunoprecipitation

Real-time quantitative polymerase chain reaction (RT-qPCR) was employed to analyze the messenger RNA (mRNA) expression levels of target genes. Total RNA was isolated by TRIzol (Thermo Fisher Scientific, United States) and reverse-transcribed into complementary DNA with a reverse transcription kit (Servicebio, China). PCR was then carried out with SYBR Green mix and specific primers (Supplementary Table 1), and the 2^-∆∆Ct method was used to quantify the relative mRNA expression of target genes. Protein expression was examined by Western blot. Protein concentrations from cells and tissues were measured with the bicinchoninic acid assay, followed by separation on 8%-10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transfer to polyvinylidene difluoride membranes. After blocking with rapid blocking buffer (Servicebio, China) for 15 minutes, membranes were incubated with primary antibodies overnight at 4 °C and then with appropriate secondary antibodies. Signals were visualized utilizing the Amersham Imager 680 system (Cytiva, United States). For co-immunoprecipitation (Co-IP), extracts from CCA cells were immunoprecipitated with specific primary antibodies (or immunoglobulin G) at 4 °C overnight. The immunoprecipitated complexes were thoroughly washed and prepared in sodium dodecyl sulfate sample buffer for subsequent western blot analysis. The antibodies used are listed in Supplementary Table 2.

Enzyme-linked immunosorbent assay and immunohistochemistry

Enzyme-linked immunosorbent assay (ELISA) assays were performed to measure the interleukin (IL)-2, interferon (IFN)-γ, tumor necrosis factor (TNF)-α, and CXCL10 levels. Supernatants from co-cultured CCA cells and T cells were collected and centrifuged to remove precipitates. Cytokine levels were quantified using commercial kits for IL-2 (SEKH-0008), IFN-γ (SEKH-0046), TNF-α (SEKH-0047) and CXCL10 (SEKH-0070) (Solarbio Science and Technology, Beijing, China). For immunohistochemistry (IHC) analysis, experimental procedures and scoring methods were performed as previously described[19]. Antibodies used for IHC are provided in Supplementary Table 2.

Animal experiments

Twelve male BALB/c nude mice (8 weeks old) were obtained from OBiO Technology and housed in a specific pathogen-free (SPF) facility with controlled humidity and temperature (22-25 °C). Animals were randomly divided into two cohorts and injected subcutaneously with 1 × 10⁶ HUCCT1 cells with stable UPB1 overexpression (HUCCT1/UPB1-OV) cells or control cells. Following injection, mice were maintained in SPF-grade animal facilities, and tumor volume, together with body weight, was measured weekly. At the sixth week, animals were sacrificed by cervical dislocation, and subcutaneous xenografts were harvested, weighed, and subjected to IHC. All animal procedures were approved and monitored by the Animal Welfare and Ethics Committee of Peking Union Medical College Hospital (approval No. XHDW-2024-32).

Statistical analysis

All statistical analyses were conducted using GraphPad Prism version 8.0.2 and R software version 4.0.1. Quantitative data are presented as mean ± SD. Data distribution and variance homogeneity were examined before statistical testing. For data conforming to a normal distribution with homogeneous variance, unpaired t-tests were applied to compare differences between two groups. For multiple group comparisons, one-way analysis of variance (ANOVA) or repeated-measures ANOVA was performed as appropriate. Correlations between variables were assessed using Spearman’s correlation coefficient. A two-sided P value < 0.05 was considered statistically significant.

RESULTS

UPB1 is down-regulated in CCA

UPB1 expression was first examined across multiple cancer types using public databases. The analysis indicated that UPB1 was down-regulated in several cancers, with a particularly pronounced reduction in CCA (Figure 1A and Supplementary Figure 1). Validation using four GEO datasets (GSE26566, GSE32225, GSE32958, and GSE76311) further confirmed that UPB1 expression was significantly lower in CCA tissues relative to normal bile duct controls (Figure 1B). RT-qPCR analysis of 20 paired fresh CCA and matched normal bile duct tissues demonstrated reduced UPB1 mRNA expression in CCA tissues (Figure 1C). In addition, IHC analysis of 84 paired CCA and adjacent normal bile duct tissues revealed lower H-scores in both iCCA and eCCA (Figure 1D-F).

Figure 1 UPB1 is down-regulated in cholangiocarcinoma. A: Low-expression of UPB1 was identified in cholangiocarcinoma (CCA) in Gene Expression Profiling Interactive Analysis database; B: Down-regulation of UPB1 was identified in CCA in four Gene Expression Omnibus datasets (GSE26566, GSE32225, GSE32958 and GSE76311); C: The messenger RNA expression levels of UPB1 were evaluated in 20 pairs of CCA and normal bile duct tissue samples; D: Representative images of UPB1 staining in CCA specimens and normal bile duct tissues; E and F: H scores of immunohistochemistry showed protein levels of UPB1 were down-regulated in both intrahepatic CCA and extrahepatic CCA tissues. T: Tumor; N: Normal; GEPIA: Gene Expression Profiling Interactive Analysis; mRNA: Messenger RNA; iCCA: Intrahepatic cholangiocarcinoma; eCCA: Extrahepatic cholangiocarcinoma.

Reduced expression of UPB1 correlates with adverse clinicopathological features and poor prognosis in CCA

Based on H-scores, 84 patients with CCA were stratified into high- and low-UPB1 expression cohorts. As presented in Supplementary Table 3 and Figure 2A, reduced UPB1 expression was correlated with adverse pathological traits, including poor histological differentiation, lymph node involvement, and advanced TNM stage. Survival analysis using Kaplan-Meier curves demonstrated that patients with reduced UPB1 expression exhibited markedly shorter OS (P = 1.687 × 10^-2) and DFS (P = 7.407× 10^-3) (Figure 2B and C). Cox multivariate analysis confirmed that reduced UPB1 expression served as an independent predictor of poor OS and DFS in CCA (Figure 2D and E and Supplementary Tables 4 and 5). Moreover, reduced UPB1 expression was also associated with poorer OS in other cancer types including cervical squamous cell carcinoma and endocervical adenocarcinoma, kidney clear cell carcinoma, kidney papillary cell carcinoma, lower grade glioma, liver hepatocellular carcinoma, LUAD, pancreatic adenocarcinoma and skin cutaneous melanoma (Supplementary Figure 1). Collectively, these findings support UPB1 as a potential prognostic predictor in CCA.

Figure 2 The down-regulation of UPB1 is associated with unfavorable clinical characteristics and outcomes in cholangiocarcinoma. A: UPB1 expression was significantly correlated with tumor differentiation, lymphnodes metastasis and tumor node metastasis stage of cholangiocarcinoma (CCA); B and C: Kaplan-Meier analyses showed CCA patients with low expression of UPB1 had inferior overall survival (OS) (B) and disease-free survival (DFS) (C) than those with high expression; D and E: Multivariate Cox regression analyses showed low expression of UPB1 was independent risk factor for OS (D) and DFS (E) of CCA patients. TNM: Tumor node metastasis; OS: Overall survival; DFS: Disease-free survival; CA19-9: Carbohydrate antigen 19-9; CI: Confidence interval.

Functional enrichment analyses and neighbor gene network of UPB1 in CCA

To investigate the relationship between UPB1 and its neighboring genes, an interaction network was generated through the GeneMANIA platform. The analysis indicated that UPB1, as the central node, was closely associated with multiple genes, including DPYS, CPS1, NIT1, VNN3, NIT2, VNN1, BTD, NADSYN1, VNN2, ARMT1, CFAP20, CYP1A2, SLC22A7, GYG2, GYG1, CFHR5, SMARCB1, ALDH8A1, ANGPTL3, and CCL16 (Figure 3A). A PPI network generated through the STRING database further revealed functional associations of UPB1 with DPYS, ALDH7A1, AGXT2, ALDH3A2, ABAT, ALDH2, ALDH9A1, CNDP1, CNDP2, and DPYD (Figure 3B). To further explore the biological functions of UPB1, 470 co-expressed genes were retrieved from the cBioPortal resource and subjected to functional enrichment analysis via DAVID 6.8. GO analysis suggested that UPB1-related genes are implicated in pathways linked to DNA replication, IL-1 production, cell growth, inflammatory response, immune response, hepatocyte differentiation, cell cycle regulation, and bile acid biosynthesis (Figure 3C). KEGG analysis demonstrated enrichment in drug metabolism, bile secretion, T cell receptor signaling, glycolysis/gluconeogenesis, and several cancer-related pathways, including peroxisome proliferators-activated receptor (PPAR) and adenosine 5’-monophosphate-activated protein kinase (Figure 3D). Finally, GSEA identified significantly enriched processes in high- and low-UPB1 expression groups, including T-cell differentiation, bile acid biosynthesis, PPAR signaling, apoptosis, focal adhesion, and immune response (Figure 3E).

Figure 3 Functional enrichment analyses and neighbor gene network of UPB1 in cholangiocarcinoma. A: GeneMANIA tool was used to show the relationships of neighbor genes of UPB1 and construct a network map; B: STRING tool was used to construct a protein-protein interaction network of UPB1; C: Gene Ontology enrichment analysis of UPB1; D: Kyoto Encyclopedia of Genes and Genomes pathway analysis of UPB1; E: Gene Set Enrichment Analysis of UPB1. BP: Biological process; CC: Cellular component; MF: Molecular function; cAMP: Cyclic adenosine monophosphate; IL: Interleukin; PPAR: Peroxisome proliferators-activated receptor; ABC: Adenosine triphosphate-binding cassette; AMPK: Adenosine 5’-monophosphate-activated protein kinase.

UPB1 overexpression suppresses proliferation and migration in CCA cells

UPB1 mRNA expression was assessed in several CCA cell lines by RT-qPCR. The analysis revealed relatively low expression in HUCCT1 cells and higher expression in QBC939 cells (Figure 4A). To investigate the biological functions of UPB1 in CCA progression, gain-of-function experiments were performed in HUCCT1/UPB1-OV, while loss-of-function experiments were conducted in QBC939 cells using two shRNAs targeting UPB1 (QBC939/UPB1-knockdown). Successful modulation of UPB1 expression was confirmed by RT-qPCR and western blot analyses (Figure 4B and C). Cellular assays were then applied to evaluate proliferation and migration. Both CCK-8 and colony formation assays demonstrated that UPB1 upregulation inhibited the proliferation of HUCCT1 cells, whereas UPB1 knockdown enhanced proliferative capacity in QBC939 cells (Figure 4D-F). Transwell and wound-healing analyses revealed that enhanced UPB1 expression impeded cell motility in HUCCT1, while UPB1 knockdown promoted migration in QBC939 cells (Figure 4G-J). Collectively, these findings indicate that UPB1 overexpression exerts inhibitory effects on the proliferation and migration of CCA cells.

Figure 4 The overexpression of UPB1 inhibits the proliferation and migration of cholangiocarcinoma cells. A: Real-time quantitative polymerase chain reaction (RT-qPCR) assays showed UPB1 was highly expressed in QBC939 cells, while relatively low-expressed in HUCCT1 cells among cholangiocarcinoma cell lines; B and C: RT-qPCR and western blot assays were used to assess the over-expression of UPB1 in HUCCT1 cells (B) and the silence of UPB1 in QBC939 cells (C); D-F: Cell counting kit-8 (D) and colony formation assays (E and F) showed that UPB1 over-expression reduced the proliferation of HUCCT1 cells, while UPB1 deletion promoted the proliferation of QBC939 cells; G-J: Transwell assays (G and H) and would healing assays (I and J) showed that UPB1 over-expression reduced the migration of HUCCT1 cells, while UPB1 deletion promoted the migration of QBC939 cells. Scale bars: 150 μm. mRNA: Messenger RNA; OV: Over-expression; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; sh-NC: Short hairpin RNA negative control; shRNA: Short hairpin RNA; CCK-8: Cell counting kit-8.

UPB1 overexpression induces cell cycle arrest and enhances apoptosis in CCA cells

Given that functional enrichment analyses indicated the involvement of UPB1 in cell cycle and apoptosis, FCM was employed to examine its impact on CCA cells. As illustrated in Figure 5A and B, UPB1 overexpression elevated apoptosis in HUCCT1 cells, whereas UPB1 knockdown reduced apoptosis in QBC939 cells. Analysis of the GEPIA dataset further revealed that UPB1 expression was negatively correlated with BCL-2 and BIRC5 in CCA (Figure 5C). Consistent with these findings, RT-qPCR and western blot analyses demonstrated that UPB1 overexpression decreased BCL-2 and BIRC5 expression in HUCCT1 cells (Figure 5D and E). Further endogenous CO-IP experiment confirmed the interaction of UPB1 with BCL-2 in HUCCT1 cells (Figure 5F). FCM also indicated that UPB1 overexpression induced G1/S phase arrest in HUCCT1 cells, while UPB1 deletion increased the proportion of QBC939 cells in the S phase (Figure 5G and H). In addition, GEPIA database analysis showed negative correlations between UPB1 expression and the key cell cycle regulators CDK2 and CCNA2 (Figure 5I). RT-qPCR and western blot analyses confirmed that UPB1 overexpression down-regulated CDK2 and CCNA2 expression in HUCCT1 cells (Figure 5J and K). The interaction of UPB1 with CDK2 was validated by CO-IP experiment in HUCCT1 cells (Figure 5L). Collectively, the findings indicate that UPB1 overexpression enhances apoptosis while triggering cell cycle arrest in CCA cells.

Figure 5 The overexpression of UPB1 arrests cell cycle and promotes apoptosis of cholangiocarcinoma cells. A and B: Flow cytometry showed that UPB1 over-expression increased the percentage of apoptotic HUCCT1 cells (A), while UPB1 knockdown reduced the percentage of apoptotic QBC939 cells (B); C: Correlations of UPB1 expression with BCL-2 and BIRC5 in Gene Expression Profiling Interactive Analysis web server; D and E: Real-time quantitative polymerase chain reaction (RT-qPCR) (D) and western blot (E) validated the expression of BCL-2 and BIRC5 was down-regulated in HUCCT1/UPB1-over-expression cells; F: Co-immunoprecipitation (Co-IP) experiment confirmed the interaction of UPB1 with BCL-2 in HUCCT1 cells; G and H: Flow cytometry showed over-expression of UPB1 triggered G1/S blockade in HUCCT1 cells, while knockdown of UPB1 increased the percentage of S phase in QBC939 cells; I: Correlations of UPB1 expression with CDK2 and CCNA2 in Gene Expression Profiling Interactive Analysis web server; J and K: RT-qPCR (J) and western blot (K) validated UPB1 over-expression reduced the expression of CDK2 and CCNA2 in HUCCT1 cells; L: CO-IP experiment confirmed the interaction of UPB1 with CDK2 in HUCCT1 cells. OV: Over-expression; sh-NC: Short hairpin RNA negative control; shRNA: Short hairpin RNA; FITC: Fluorescein isothiocyanate; mRNA: Messenger RNA; TPM: Transcripts per million; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; IP: Immunoprecipitation; IgG: Immunoglobulin G.

UPB1 participates in the tumor immune infiltration of CCA

Immune evasion within the TME is a fundamental hallmark of cancer[5]. Since enrichment analyses indicated potential involvement of UPB1 in immune regulation, bioinformatics analyses and co-culture experiments were conducted to clarify its role in tumor immune infiltration in CCA. As depicted in Figure 6A, UPB1 expression displayed predominantly positive correlations with TILs in the TISIDB database. Strongest positive correlations were observed with activated CD4⁺ T cells, neutrophils, and γδ T cells (Figure 6B). Associations between UPB1 expression and immunostimulators, immunoinhibitors, and chemokines are presented in Supplementary Figure 2. In addition, UPB1 expression showed positive correlations with T-cell surface markers (CD2, CD8A, and STAT6), while negative correlations were observed with the M2 macrophage marker CD163 and immune checkpoint molecules such as TIM3, LMTK3, and VISTA (Figure 6C and D). To further validate these findings, HUCCT1 cells were co-cultured with T cells in Transwell chambers, and culture supernatants were analyzed (Figure 6E). Western blot analysis revealed that UPB1-overexpressing HUCCT1 cells co-cultured with T cells displayed elevated levels of T-cell activation markers CD25, CD44, and CD69 relative to controls (Figure 6F). ELISA assays further demonstrated that IL-2, IFN-γ, TNF-α, and CXCL10 secretion were elevated in the UPB1-overexpression group (Figure 6G). Collectively, these in vitro findings indicate that UPB1 overexpression may promote T-cell activation and enhance the release of immune factors, indicating that UPB1 may be associated with tumor immune infiltration in CCA.

Figure 6 UPB1 participates in tumor immune infiltration of cholangiocarcinoma. A: Correlations between UPB1 expression and tumor-infiltrating lymphocytes (TILs) in TISIDB web server; B: Top 3 TILs with greatest positive Spearman’s correlation with UPB1; C and D: Correlations between UPB1 expression and gene markers of TILs and other immune checkpoints; E: Example diagram for co-culture of HUCCT cells and T cells; F: Western blot analysis on the expressions of T cell activation markers, CD25, CD44 and CD69; G: Elisa assay on the levels of interleukin-2, interferon-γ, tumor necrosis factor-α and CXCL10. CD: Cluster of differentiation; Th: T helper; Treg: Regulatory T cells; NK: Natural killer; MDSC: Myeloid-derived suppressor cells; NKT: Natural killer T cells; DC: Dendritic cells; pDC: Plasmacytoid dendritic cells; iDC: Immature dendritic cells; CECS: Conventional endometrial carcinosarcoma; CHOL: Cholangiocarcinoma; COAD: Colon adenocarcinoma; ESCA: Esophageal carcinoma; GBM: Glioblastoma multiforme; HNSC: Head and neck squamous cell carcinoma; KICH: Kidney chromophobe; KIRC: Kidney renal clear cell carcinoma; KIRP: Kidney renal papillary cell carcinoma; LGG: Lower grade glioma; LIHC: Liver hepatocellular carcinoma; LUAD: Lung adenocarcinoma; MESO: Mesothelioma; OV: Ovarian serous cystadenocarcinoma; PAAD: Pancreatic adenocarcinoma; PCPG: Pheochromocytoma and paraganglioma; PRAD: Prostate adenocarcinoma; READ: Rectum adenocarcinoma; SARC: Sarcoma; SKCM: Skin cutaneous melanoma; STAD: Stomach adenocarcinoma; TGCT: Testicular germ cell tumor; THCA: Thyroid papillary carcinoma; UCEC: Uterine corpus endometrial carcinoma; UCS: Uterine carcinosarcoma; UVM: Uveal melanoma; OV: Over-expression; GAPDH: Glyceraldehyde-3-phosphate dehydrogenase; IL: Interleukin; IFN: Interferon; TNF: Tumor necrosis factor.

The overexpression of UPB1 inhibits tumor growth in vivo

To evaluate its role in tumor progression, HUCCT1/UPB1-OV and control cells were subcutaneously implanted into BALB/c nude mice (n = 6) to generate xenografts (Figure 7A). Mice in the UPB1-OV group exhibited a markedly slower increase in tumor volume compared with controls (Figure 7B and C). At the endpoint, tumor weight was also lower in the UPB1-OV cohort (Figure 7D), confirming a modest tumor-suppressive effect of UPB1 in vivo. Conversely the deletion of UPB1 significantly promoted tumor growth in vivo (Supplementary Figure 3). IHC analysis further revealed reduced expression of Ki-67, E-cadherin, and CD31 in xenograft tissues derived from the UPB1-OV group, while TUNEL and CXCL10 staining were enhanced (Figure 7E), which was consistent with the increased CXCL10 secretion observed in our in vitro co-culture system (Figure 6G). Collectively, these findings demonstrate that UPB1 overexpression inhibits tumor growth in vivo and supports its role as a suppressor in CCA.

Figure 7 The overexpression of UPB1 moderately inhibits tumor growth in vivo. A and B: BALB/c-nudes (n = 6 per group) were sacrificed 6 weeks after the injection and tumors dissected from respective groups were shown; C: Tumor growth curves after the injection of HUCCT1/UPB1-over-expression (OV) and control cells. Tumor volume was calculated every week; D: Tumor weight was measured in UPB1-OV and control groups; E: Immunohistochemistry staining of Ki67, E-Cadherin, TUNEL, cluster of differentiation 34, and CXCL10 in tumors from UPB1-OV and control groups; F: A schematic diagram integrated findings from in vitro and in vivo experiments, for a novel regulatory mechanism that UPB1 induced apoptosis by BCL-2/BIRC5 axis, caused G1/S phase arrest through CDK2/CCNA2 pathway and also contributed to tumor immune infiltration by enhancing T-cell activation and promoting the release of immune factors in cholangiocarcinoma. Scale bars: 200 μm. OV: Over-expression; CD: Cluster of differentiation; E-Cad: E-Cadherin; IL: Interleukin; IFN: Interferon; TNF: Tumor necrosis factor.

DISCUSSION

The discovery of reliable biomarkers and clarification of the biological processes driving carcinogenesis are critical for advancing therapeutic approaches and improving survival outcomes. Previous investigations have identified UPB1 as a novel circulating signature associated with DFS in HCC through integrative bioinformatics analyses[20]. In addition, UPB1 variants have been linked to increased toxicity and altered outcomes of fluoropyrimidine-based chemotherapy, suggesting a potential role in treatment response[21,22]. Nevertheless, the prognostic value of UPB1 may vary across different cancer types, reflecting potential tissue-specific functions. In this study, down-regulation of UPB1 in CCA was found to be associated with unfavorable clinical features, including poor differentiation, lymph node involvement, and advanced TNM stage. Patients with diminished UPB1 levels showed reduced overall and DFS, underscoring its potential as an independent prognostic indicator in CCA. In vitro studies demonstrated that UPB1 overexpression enhanced apoptosis and triggered G1/S cell cycle arrest in CCA cells. Reduced expression of BCL-2 and BIRC5, as well as down-regulation of CDK2 and CCNA2, were observed in UPB1-overexpressed cells. Xenograft experiments further validated that UPB1 overexpression leads to a moderate reduction in tumor growth in vivo. Moreover, UPB1 expression showed positive correlations with TILs, particularly T-cell surface markers. Co-culture experiments with CCA and T cells revealed that UPB1 overexpression enhanced the presence of T-cell activation markers, including CD25, CD44, and CD69, and increased the secretion of immune factors like IL-2, IFN-γ, TNF-α, and CXCL10. Overall, these findings identify UPB1 as a prognostic biomarker and tumor suppressor associated with immune infiltration in CCA (Figure 7F).

Excessive proliferation of cancer cells is a fundamental characteristic of malignant tumors, largely driven by dysregulated cell cycle control and impaired apoptotic mechanisms[23]. Since enrichment analysis suggested a role for UPB1 in cell cycle control and apoptosis, its influence on CCA cells was examined by FCM. The results demonstrated that UPB1 overexpression promoted apoptosis by regulating BCL-2 and BIRC5, while inducing G1/S phase arrest through modulation of the CDK2/CCNA2 pathway. Cell cycle checkpoints, primarily governed by cyclins and cyclin-dependent kinases, function as DNA surveillance systems that halt progression and trigger cell death when DNA damage cannot be repaired[24]. CDK2/CCNA2 is a wellestablished regulator of G1/S transition[25-27]. In the present study, UPB1 overexpression led to reduced expression of CDK2 and CCNA2, accounting for the observed G1/S arrest in UPB1-OV cells. In addition, previous research demonstrated that UPB1 overexpression in CCA cells elevated the levels of the apoptotic regulators BCL-2 and BIRC5. Within apoptotic signaling, BCL-2 prevents mitochondrial cytochrome c release, thereby inhibiting caspase activation, which represents a central mechanism underlying both programmed cell death and chemotherapy resistance[28,29]. While our Co-IP experiments confirm that UPB1 associates with BCL-2 and CDK2 in CCA cells, it remains to be determined whether these interactions are direct or mediated by other components of the protein complex. Future studies employing in vitro binding assays or rescue experiments will be necessary to establish direct physical interaction and causal direction. Notably, UPB1 is an enzyme involved in pyrimidine catabolism. Although metabolic intermediates were not measured here, it is plausible that UPB1 loss in CCA disrupts pyrimidine nucleotide homeostasis, thereby promoting proliferation and resisting apoptosis. Whether such metabolic alterations directly mediate the tumor-suppressive effects of UPB1 in CCA remains to be determined, but this represents a plausible mechanistic link worth investigating.

Tumor immune infiltration is recognized as a hallmark of most cancers, including CCA, and has emerged as an effective and widely explored therapeutic target[30]. Consistent with previous findings[12], the present study demonstrated that UPB1 expression was strongly linked to immune cell presence in the TME. In LUAD, large-scale bioinformatic evaluations showed positive correlations between UPB1 expression and the abundance of macrophages, B lymphocytes, and CD8⁺ T cells[9]. Another study using immunohistochemical detection in a cohort of 28 LUAD patients further confirmed a positive association between UPB1 and CD8 expression and established a CD8⁺ T cell related risk model incorporating UPB1 to predict survival in stage III LUAD[12]. However, functional studies directly addressing the effects of UPB1 on TILs remain limited. In the present work, UPB1 expression showed strong positive correlations with T-cell abundance in CCA. Co-culture experiments with CCA cells and T cells further revealed that UPB1 overexpression enhanced T-cell activation markers and stimulated cytokine secretion. Circulating markers such as CD25, CD44, and CD69 are well-established indicators of T-cell activation[31-33], while cytokines including IL-2, IFN-γ, TNF-α, and CXCL10 promote the functional activity of TILs, including CD8⁺ T and natural killer cells, strengthening cytotoxicity[34,35]. Further investigations are warranted to employ patient-derived organoids co-cultured with autologous immune cells to confirm these mechanisms in a patient-specific context and to explore their potential for predicting immunotherapy responses.

This study provides the earliest evidence that UPB1 serves as a prognostic indicator in CCA and influences both tumor cell aggressiveness and immune infiltration. This potential dual role of UPB1 may delineate a distinctive regulatory network in CCA pathogenesis. Unlike classic suppressors whose networks are often centered on intracellular signaling cascades, the UPB1-associated network encompasses pathways like T-cell receptor signaling and chemokine activity. Functionally, this suggests that loss of UPB1 might confer a dual advantage to tumor cells: Evading intrinsic growth control while simultaneously fostering an immunosuppressive milieu. Therefore, restoring UPB1 function could represent a strategy that simultaneously targets the tumor cell’s core vulnerabilities and its supportive immune niche, offering a novel therapeutic axis distinct from targeting conventional suppressor pathways. Our findings align with recent reports linking UPB1 to immune infiltration in LUAD[9,12], but provide the first functional evidence in CCA using co-culture experiments. Nonetheless, several limitations should be acknowledged. First, while our in vitro co-culture experiments firmly establish the role of UPB1 in modulating T-cell activation, the current study utilized T-cell deficient nude mice for in vivo tumor growth assays, which precludes direct observation of UPB1-mediated immune surveillance in vivo. Given the established importance of T-cell responses in CCA progression and immunotherapy response[5], immunocompetent models are essential to determine whether UPB1 influences anti-tumor immunity in the intact TME. Future studies employing immunocompetent mouse models or humanized mouse models are essential to fully elucidate the impact of UPB1 on the tumor immune microenvironment and to validate its potential as a target for immunotherapy. Second, future multi-center prospective studies are needed to validate the generalizability of UPB1’s prognostic value. Third, our study did not identify the upstream cause of UPB1 down-regulation in CCA. Promoter hypermethylation and transcriptional repression by oncogenic pathways are common mechanisms driving tumor suppressor silencing in cancers[36]; whether such mechanisms account for reduced UPB1 expression in CCA remains an open question. Future work is needed to investigate whether reduced UPB1 expression is driven by epigenetic mechanisms or metabolic regulation, both of which play pivotal roles in tumor initiation and progression.

CONCLUSION

In summary, overexpression of UPB1 enhanced apoptosis and triggered G1/S cell cycle arrest in CCA cells. Based on in vitro co-culture and bioinformatics data, UPB1 may participate in tumor immune infiltration in CCA with enhanced T-cell activation and increased immune factor release. Collectively, the findings support a tumorsuppressive role of UPB1 in CCA and suggest UPB1 as a candidate for further investigation.