3.1 Gene expression analysis of EBNA1BP2

In our analysis using the TIMER2.0 database, we observed that the expression level of EBNA1BP2 was generally higher in tumor samples compared to adjacent normal tissues in most of cancer types. Specifically, we found higher expression levels of EBNA1BP2 in bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), cholangiocarcinoma (CHOL), colon adenocarcinoma (COAD), esophageal carcinoma (ESCA), head and neck squamous cell carcinoma (HNSC), kidney renal clear cell carcinoma (KIRC), liver hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), prostate adenocarcinoma (PRAD), rectum adenocarcinoma (READ), stomach adenocarcinoma (STAD), thyroid carcinoma (THCA), and uterine corpus endometrial carcinoma (UCEC) (Fig. 1A). However, in kidney chromophobe (KICH), we observed the opposite trend, with lower expression of EBNA1BP2 in tumor samples compared to normal tissues. To further validate our findings, we combined data from both TCGA and GTEx databases. In this analysis (Fig. 1B), we found higher expression of EBNA1BP2 in diffuse large B-cell lymphoma (DLBC), ovarian serous cystadenocarcinoma (OV), and thymoma (THYM) tumors compared to adjacent normal controls. In contrast, no significant differential expression was observed in adrenocortical carcinoma (ACC), brain lower grade glioma (LGG), sarcoma (SARC), and uterine carcinosarcoma (UCS) tumors (Supplementary Figure S1). Overall, our analysis indicates that EBNA1BP2 is highly expressed in most tumor types, suggesting its potential role in cancer development and progression.

Fig. 1

Expression and protein levels of EBNA1BP2 in pan-cancer. A Expression level of EBNA1BP2 in human tumors from TIMER2.0. B Box plot representation the expression level of EBNA1BP2 in DLBC, LAML, OV and THYM from GTEx and TCGA database. C Total protein level of EBNA1BP2 in BRCA, OV, COAD, KIRC, LUAD, HNSC, PAAD, GBM and LIHC were analyzed using CPTAC. D GEPIA2.0 was used to compare EBNA1BP2 expression levels in different pathological stages of KICH, KIRC and LIHC. *p < 0.05; **p < 0.01; ***p < 0.001. EBNA1BP2: EB nuclear antigen 1 binding protein 2; GTEx: Genotype-Tissue Expression; TCGA: The Cancer Genome Atlas; CPTAC: Clinical Proteomic Tumor Analysis Consortium; DLBC: lymphoid neoplasm diffuse large B-cell lymphoma; LAML: acute myeloid leukemia; OV: ovarian serous cystadenocarcinoma; THYM: thymoma; BRCA: breast invasive carcinoma; COAD: colon adenocarcinoma; KIRC, kidney renal clear cell carcinoma; LUAD: lung adenocarcinoma; HNSC: head and neck squamous cell carcinoma; PAAD: pancreatic cancer; GBM: polymorphous glioblastoma; LIHC, liver hepatocellular carcinoma; KICH, kidney chromophobe

To analyze EBNA1BP2 expression at the protein level, we utilized the National Cancer Institute's CPTAC tool. Our analysis revealed that the total protein expression of EBNA1BP2 significantly increased in BRCA, OV, COAD, KIRC, LUAD, HNSC, pancreatic adenocarcinoma (PAAD), polymorphous glioblastoma (GBM), and LIHC (Fig. 1C). Furthermore, we examined the relationship between EBNA1BP2 expression and clinical tumor pathological stages using the GEPIA2.0 tool. Our analysis indicated that the expression level of EBNA1BP2 in KICH, KIRC, and LIHC showed significant differences across different pathological stages (Fig. 1D; Supplementary Figure S2).

In order to further validate the expression of EBNA1BP2, IHC results of the tumors were acquired using the HPA database. The IHC staining of EBNA1BP2 was found to be moderately or strongly expressed in BRCA, LICH, and LUSC tumors. This analysis was consistent with previous EBNA1BP2 gene expression results from the TCGA database (Fig. 2A–C).

Fig. 2

Comparison of EBNA1BP2 expression in normal tissue and tumor tissue. A–C The expression level of EBNA1BP2 obtained by UALCAN and the corresponding immunohistochemical images obtained by HAP platform both showed up-regulated expression of EBNA1BP2 in breast, liver and lung derived tumor tissues. UALCAN: The University of ALabama at Birmingham CANcer data analysis Portal; HPA: Human Protein Atlas

3.2 Survival analysis data

Next, we explored the relationship between the expression of EBNA1BP2 and the prognosis of survival. The result of GEPIA2.0 showed that low levels of EBNA1BP2 expression in ACC, BLCA, LGG, LIHC, MESO, SARC, UCS were significantly associated with longer OS (Fig. 3A). In ACC, HNSC, KIRP, LGG, PRAD, SARC and UCS, low EBNA1BP2 expression was also associated with better DFS (Fig. 3B).

Fig. 3

The relationship between the expression level of EBNA1BP2 and prognosis of pan-cancer. A, B Survival charts and Kaplan–Meier curves show the relationship between the expression level of EBNA1BP2 and OS (A) and DFS (B) in human tumors were obtained by GEPIA2.0. OS: overall survival; DFS: disease-free survival; GEPIA2.0: Gene Expression Profiling Interactive Analysis, version 2

As shown in Supplementary Figure S3, our analysis revealed that lower expression of EBNA1BP2 in BLCA, LIHC, LUAD, SARC, and UCEC was associated with improved OS. However, the results were contrary in KIRC, OV, PCPG, STAD, and THCA. Specifically, the combined analysis of GEPIA2.0 and Kaplan–Meier plotter tools showed that low EBNA1BP2 expression was significantly associated with longer OS in BLCA, LIHC, and SARC. These findings were primarily derived from the TCGA database. Additionally, we further assessed the OS for EBNA1BP2 using the PrognoScan website, which mainly extracted data from the GEO database. The results demonstrated that EBNA1BP2 expression was significantly associated with bladder, brain, and lung cancers, and surprisingly, low EBNA1BP2 expression was also related to better OS (Supplementary Figure S4A-C). These findings suggest that EBNA1BP2 may serve as a potential prognostic marker for a wide variety of cancer types.

3.3 Genetic alteration landscape analysis

The EBNA1BP2 gene mutation has been shown to affect cellular functions [20]. To analyze the mutation status of the EBNA1BP2 gene in various types of tumors, we utilized the cBioPortal platform. The results revealed that in ovarian cancer (OV), the primary type of alteration was "amplification," with the highest frequency of EBNA1BP2 alteration (> 5%). UCEC had the highest incidence of "mutation," with a frequency of approximately 2% of cases (Fig. 4A). Figure 4B illustrates other mutations and their locations within the EBNA1BP2 gene. We observed that missense mutations were the most common type of mutation in EBNA1BP2. Among these, the G291 = /K291N/X291_splice mutation was the most frequently mutated region of the EBNA1BP2 protein. Figure 4C displays the gene site visualized in the 3D structure of the EBNA1BP2 protein. Furthermore, we analyzed the potential association between EBNA1BP2 gene alterations and pan-cancer survival prognosis. However, we found that alterations in the EBNA1BP2 gene did not significantly impact patient prognosis (Supplementary Figure S5). Further verification may require additional patient clinical data in the future.

Fig. 4

EBNA1BP2 gene mutation in different cancers. The altered frequencies of different mutation types (A) and mutation sites (B) in pan-cancer were shown using cBioPortal. C The 3D protein structure of EBNA1BP2 was shown. cBioPortal, The cBio Cancer Genomics Portal

3.4 Methylation level analysis

Abnormal methylation of gene promoter regions is frequently implicated as a contributing factor in carcinogenesis [21]. To investigate the methylation levels of the EBNA1BP2 gene in tumor and normal tissues, we utilized the UALCAN tool. The results demonstrated that the promoter methylation of the EBNA1BP2 gene was significantly higher in normal tissues compared to tumor tissues. However, in KIRC and KIRP, the methylation of the EBNA1BP2 promoter was significantly increased (Fig. 5; Supplementary Figure S6). These findings suggest that altered promoter methylation may play a role in the transcriptional expression of EBNA1BP2.

Fig. 5

Comparison of differences in promoter methylation levels of EBNA1BP2 in cancer. EBNA1BP2 methylation values of normal tissues and primary tumor tissues were analyzed by UALCAN tool

3.5 Immune infiltration analysis data

Studies have demonstrated that immune cell infiltration is closely associated with various tumor behaviors, including tumor occurrence and development [22, 23]. Therefore, we utilized multiple algorithms such as TIMER, EPIC, QUANTISEQ, XCELL, MCPCOUNTER, CIBERSORT, CIBERSORT-ABS, and TIDE to investigate the correlation between EBNA1BP2 expression and immune cell infiltration in pan-cancer. In LUSC, we observed an inverse correlation between EBNA1BP2 expression and B-cell infiltration (Fig. 6A). Additionally, in BRCA, STAD, TGCT, and THCA, there was an inverse correlation between cancer-associated fibroblast infiltration and EBNA1BP2 expression (Fig. 6B). In UVM, EBNA1BP2 expression was positively correlated with CD8+ T cell infiltration (Fig. 6C). Furthermore, we found that EBNA1BP2 expression was positively correlated with neutrophils in KIRC and negatively correlated with monocytes in THCA (Supplementary Figure S7; Supplementary Figure S8A-E). These findings suggest that EBNA1BP2 may have potential as a tumor immune-related biomarker in the future.

Fig. 6

The correlation between EBNA1BP2 expression level and immune cells in tumors. A–C TIMER2.0 database was used to analyze the relationship between EBNA1BP2 expression and immune infiltration of B cells (A), cancer-associated fibroblasts (B) and T cells CD8+ (C). TIMER, EPIC, TIDE, QUANTISEQ, CIBERSORT, CIBERSORT-ABS, XCELL, MCPCOUNTER and other algorithms are used for analysis. Red for positive correlation (0–1) and blue negative correlation (− 1 to 0). A correlation of p < 0.05 was considered statistically significant. A cross indicates that the correlation is not significant. TIMER2.0: Tumor Immune Estimation Resource, version 2

3.6 Single cell sequencing data analysis

To further validate the potential functions of the candidate genes at the level of the single cell, we used the CancerSEA tool to investigate the correlation of EBNA1BP2 gene expression with the function of cancer cells in pan-cancer. EBNA1BP2 expression was positively correlated with angiogenesis, differentiation, and inflammation in retinoblastoma (RB). Conversely, it was inversely correlated with cell cycle, DNA damage response, and DNA repair response. In uveal melanoma (UM), EBNA1BP2 expression showed negative correlations with almost all tumor biological behaviors, including apoptosis, DNA repair response, invasion, and metastasis (Fig. 7A). Furthermore, Fig. 7B shows significant correlations between EBNA1BP2 expression and angiogenesis differentiation and DNA repair in RB, DNA repair and DNA damage in UM, and quiescence in LUAD. And then, EBNA1BP2 single-cell expression profile of RB, UM and LUAD by T-SNE plots is also shown in Fig. 7C.

Fig. 7

EBNA1BP2 gene expression at the single-cell level. A, B Using CancerSEA tools analysis EBNA1BP2 expression and tumor in the relationship between different functional status. C The single-cell expression profile of EBNA1BP2 in RB, UM and LUAD was shown by T-SNE plot. *p < 0.05; **p < 0.01; ***p < 0.001. RB: retinoblastoma; UM: uveal melanoma; LUAD: lung adenocarcinoma

3.7 Gene enrichment analysis

Finally, we combined all of the tumor expression data from the TCGA from the GEPIA2.0 tool to obtain the top 100 genes associated with the expression of EBNA1BP2 (Supplementary Table S1). Subsequently, functional enrichment analysis was performed to evaluate these potential mechanisms. Using the BioGRID network tool, we identified 11 molecules that interact with EBNA1BP2 (Fig. 8A). Notably, EBNA1BP2 expression showed a high correlation with Homo sapiens peptidylprolyl isomerase H (PPIH), diphthamide biosynthesis 2 (DPH2), Homo sapiens Y-box in most cancers binding protein 1 (YBX1), Homo sapiens MRT4 homolog, ribosome maturation factor (MRTO4), Homo sapiens mediator complex subunit 8 (MED8), and Homo sapiens cell division cycle 20 (CDC20) (Fig. 8B, C). Furthermore, GO and KEGG enrichment analysis revealed that the genes associated with EBNA1BP2 expression were primarily related to the nucleoplasm and RNA binding pathways (Fig. 8D). These findings suggest that EBNA1BP2 may play a role in tumorigenesis and development through these pathways.

Fig. 8

Functional enrichment and pathway analysis of EBNA1BP2-related genes. A EBNA1BP2 related gene is available by the BioGRID website and 11 proteins have been shown. B The top six genes most related to EBNA1BP2 were obtained using GEPIA2.0, which were PPIH, DPH2, YBX1, MRTO4, MED8 and CDC20. p-value < 0.001. C Heat map results confirmed EBNA1BP2 gene expression and six (PPIH DPH2, YBX1, MRTO4, MED8 and CDC20) in pan-carcinoma were positively correlated. D GO and KEGG enrichment analysis of EBNA1BP2-related genes. BioGRID: Biological General Repository for Interaction Datasets; PPIH: Homo sapiens peptidylprolyl isomerase H; DPH2: diphthamide biosynthesis 2; YBX1: Homo sapiens Y-box in most cancers binding protein 1; MRTO4: Homo sapiens MRT4 homolog, ribosome maturation factor; MED8: Homo sapiens mediator complex subunit 8; CDC20: Homo sapiens cell division cycle 20; GO: Gene Ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes

3.8 The expression of genes in cancer and normal tissue

In addition, we detected the expression of EBNA1BP2 in clinical samples of LUAD and normal lung tissue using RT-PCR method. The results showed that the expression of the EBNA1BP2 was upregulated in lung cancer tissues compared to normal tissues (Supplementary Figure S9).