3.1. Differential Expression of ZBTB11 in Tumor Versus Normal Tissue Samples
We used TIMER2.0 software to compare the expression levels of ZBTB11 in 33 cancer tissues and their corresponding adjacent normal tissues. The results, shown in Fig. 1A, revealed significant differences in the transcription levels of ZBTB11 between tumor and normal tissues in 13 cancer types. Specifically, increased ZBTB11 expression was observed in cholangio carcinoma (CHOL), colon adenocarcinoma (COAD), esophageal carcinoma (ESCA), head and neck squamous cell carcinoma (HNSC), liver hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), and stomach adenocarcinoma (STAD). Conversely, reduced ZBTB11 expression was found in glioblastoma multiforme (GBM), kidney chromophobe (KICH), kidney renal papillary cell carcinoma (KIRP), thyroid carcinoma (THCA), and uterine corpus endometrial carcinoma (UCEC) compared to normal tissues. Since the TCGA database could not provide the normal tissues of some tumors for comparison, we extracted expression data from Genotype-Tissue Expression (GTEx) and found that ZBTB11 was significantly upregulated in CHOL, lymphoid neoplasm diffuse large B-cell lymphoma (DLBC), and thymoma (THYM) tissues compared to normal tissues (Fig. 1B). Moreover, analyses of tumor tissues from 33 cancer types in The Cancer Genome Atlas (TCGA) database showed that ZBTB11 mRNA expression levels were tumor-specific, with the highest expression in acute myeloid leukemia (LAML) (Fig. 1C).
To investigate the mRNA and protein expression profiles of ZBTB11 in both normal and cancerous human tissues, we analyzed the ZBTB11 expression data for normal tissues using the HPA database. An overview of the ZBTB11 mRNA and protein expression data revealed a low tissue specificity for ZBTB11 distribution (Fig. S1). In addition, IHC results from the HPA database showed higher staining intensity of ZBTB11 in many cancers, particularly lung, ovarian, pancreatic, and urothelial cancers (Fig. 1D, S2). In summary, ZBTB11 exhibited overexpression in most cancers.
3.2. Pan-cancer Analysis of the Relationship between ZBTB11 Expression and Clinicopathology
We investigated the association between ZBTB11 and clinicopathologic characteristics using the UALCAN tool to assess the expression pattern of ZBTB11 in normal tissues and stage I, II, III, and IV tumor tissues in various cancers. ZBTB11 expression was significantly upregulated during disease progression in several cancers including CHOL, COAD, ESCA, HNSC, LIHC, LUAD, LUSC, and STAD. Conversely, ZBTB11 expression was significantly downregulated during disease progression in KICH, THCA and UCEC (Fig. 2). We also examined the effect of ZBTB11 expression in different cancer subtypes and grades (Fig. S3).
3.3. Prognostic significances of ZBTB11 in different cancers
ZBTB11 is aberrantly expressed in many tumors, prompting us to investigate the prognostic relevance of ZBTB11 in cancer. Data from GEPIA2 revealed correlations between ZBTB11 expression and OS in four cancer types. Specifically, increased ZBTB11 expression correlated with a favorable prognosis in CHOL and kidney renal clear cell carcinoma (KIRC), while the opposite trend was observed in other cancers such as LIHC and prostate adenocarcinoma (PRAD) (Fig. 3A). Additionally, analyses using TISIDB indicated that high ZBTB11 expression was associated with longer OS in GBM and KIRC (Fig. 3B). Furthermore, K-M Plotter demonstrated clear survival differences between high and low ZBTB11 expression groups across multiple cancers (Fig. 3C). The combined results from these databases underscored the predictive value of high ZBTB11 expression for favorable overall survival in KIRC (Fig. 3D). Finally, Cox regression analysis was performed to evaluate the relationship between ZBTB11 expression and survival outcomes of cancer patients, including OS, disease-specific survival (DSS), disease-free interval (DFI), and progression-free interval (PFI). Specifically, the results of Cox regression analysis showed that ZBTB11 expression levels were associated with OS, DSS and PFI in LIHC (Fig. S4). These findings indicated that ZBTB11 was a risk factor in LIHC, brain lower-grade glioma (LGG), and breast invasive carcinoma (BRCA), while it was a protective factor for KIRC, skin cutaneous melanoma (SKCM), and rectum adenocarcinoma (READ).
3.4. Methylation of ZBTB11 in cancers
Abnormal DNA methylation associated with increased cancer risk [41]. The methylation levels of ZBTB11 in pan-cancer tissues were compared with those in normal tissues using UALCAN. ZBTB11 methylation levels were found to be significantly higher in BRCA (P = 2.424E-03), KIRC (P = 5.460E-09), KIRP (P = 7.772E-16), and THCA (P = 1.879E-03) tissues than in normal tissues, and significantly lower in bladder urothelial carcinoma (BLCA) (P = 8.105E-15), COAD (P = 1.624E-12), HNSC (P = 1.630E-12), LIHC (P = 1.883E-08), LUAD (P = 1.624E-12), LUSC (P = 1.624E-12), PRAD (P < 1E-12), READ (P = 1.051E-06), sarcoma (SARC) (P = 7.038E-03), testicular germ cell tumors (TGCT) (P = 2.337E-08), and UCEC (P = 1.110E-16) tissues compared to normal tissues (Fig. 4A). Additionally, the GSCA database was utilized to explore survival differences between high and low methylation of ZBTB11 in specific cancers. Patients with higher ZBTB11 methylation in LIHC had better DSS and OS, while patients with lower methylation in BLCA had a better prognosis for OS and progression-free survival (PFS). Conversely, patients with higher ZBTB11 methylation in LGG had a worse prognosis for DSS, OS, and PFS (Fig. 4B). Furthermore, MethSurv was employed to analyze the effect of ZBTB11 methylation levels on patient survival and prognosis in different types of cancer. It was found that high methylation levels of ZBTB11 led to poor survival and prognosis in patients with COAD, LAML, LIHC, LUSC, mesothelioma (MESO) and UCEC (Fig. S5).
3.5. The genetic and epigenetic features of ZBTB11 in cancers
GSCA was used to analyze the SNV and CNV of ZBTB11 in various cancers. UCEC, uterine carcinosarcoma (UCS), and STAD exhibited the highest percentage of SNVs with 6.97, 3.51 and 3.42, respectively. Predominant variation types were missense mutations and single nucleotide polymorphisms (SNPs), most commonly C > T (Fig. 5A, B). For CNVs, heterozygous amplification variations of ZBTB11 were most common in LUSC, cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), and ovarian serous cystadenocarcinoma (OV), with amplification percentages of 56.29%, 46.78%, and 45.77%, respectively (Fig. S6A). Homozygous amplifications of ZBTB11 were most common in LUSC, CESC, and ESCA, with amplification percentages of 8.58%, 4.41%, and 4.35%, respectively (Fig. S6B). Additionally, the percentage and contribution of CNV to pan-cancer ZBTB11 expression were assessed, revealing a higher percentage of CNV in LUSC, CESC, OV, HNSC, and BLCA, with heterozygous amplifications accounting for the largest proportion (Fig. 5C). We utilized the cBioPortal database to analyze the genetic variants of ZBTB11 in pan-cancer. ZBTB11 mutations accounted for the largest proportion of all mutation types, with the highest frequency in UCEC, UCS, and STAD at 6.81%, 3.51%, and 3.41%, respectively (Fig. 5D). Analysis from the TIMER2.0 database revealed that UCEC (37/531), UCS (2/57), and STAD (15/439) had the highest ZBTB11 gene mutation rates (Fig. S6C). Furthermore, analysis of the cBioPortal database indicated that missense mutations were the predominant type of ZBTB11 gene mutation in cancers (Fig. 5E). Copy number alterations in ZBTB11 were most commonly observed as amplifications, gain-of-function, and diploidy (Fig. S6D). Among the ZBTB11 group with alterations, ALOX12P1, SMPD4P1, TXN, TTN, PIK3CA, EPHA6, SENP7, SI, TP53, and NXPE3 genes had the highest frequency of alterations (Fig. S6E). We recognized the importance of the mRNA modification parameter in epigenetics, which is involved in post-transcriptional gene regulation [42]. As depicted in Fig. S7, ZBTB11 expression exhibited a positive correlation with most m1A, m5C, and m6A methylation in pan-cancer tissues.
3.6. Interactions between ZBTB11 and immune infiltration in cancers
The interaction between tumors and the immune system plays a critical role in the development, progression and treatment of cancer [43]. Therefore, elucidating the interplay between tumor and immune cells would aid both in predicting immunotherapy responses and in the development of novel immunotherapy targets [44]. We used the TISIDB web portal, which integrates multiple heterogeneous data types related to tumor-immune system interactions, to identify potential correlations between ZBTB11 expression and immune cells, immune inhibitors, immune stimulators, MHC, chemokines, and receptors (Fig. 6A-F). In the context of LIHC, Fig. 6G presented compelling evidence of an inverse correlation between ZBTB11 levels and the presence of gamma delta T cells, activated CD8 + T cells, CD56bright natural killer cells, monocyte cells, and CD56dim natural killer cells. Additionally, we utilized the TISIDB database to examine the associations between ZBTB11 expression and immune cells infiltration. ZBTB11 demonstrated strong correlations with various Immunoinhibitors and Immunostimulators, including CD244, TGFBR1, KDR, PVRL2, LGALS9, TNFRSF8, C10orf54, TNFRSF14, TNFRSF18, and IL6R (Fig. 6H, I). Furthermore, a significant correlation was observed between ZBTB11 and certain chemokines, specifically CCL19, XCL2, CCL14, CCL3, and CCL5. Finally, ZBTB11 also exhibited a strong association with various chemokine receptors, such as CX3CR1, CXCR6, CCR10, CCR4, and CXCR3 (Fig. 6J, K).
3.7. Functional states analysis of ZBTB11 at single-cell levels
We utilized CancerSEA to examine the functional states of ZBTB11 at the single-cell level in different types of cancer. The results revealed that ZBTB11 was positively correlated with angiogenesis, differentiation, inflammation, and stemness, and negatively correlated with apoptosis, the cell cycle, DNA damage, DNA repair, epithelial–mesenchymal transition (EMT), hypoxia, invasion, metastasis, and proliferation (Fig. 7A). Subsequently, we further examined the association between ZBTB11 and specific cancer types. ZBTB11 was found to be positively correlated with angiogenesis in AML, with differentiation and stemness in glioma, with stemness in high-grade glioma (HGG), and with angiogenesis, differentiation and inflammation in retinoblastoma (RB). Conversely, ZBTB11 demonstrated a negative correlation with invasion, metastasis, DNA repair, apoptosis, DNA damage, proliferation, EMT, hypoxia, and the cell cycle in BRCA; with invasion, metastasis, and EMT in renal cell carcinoma (RCC); with DNA repair, the cell cycle, and DNA damage in RB; and with DNA repair, DNA damage, and apoptosis in uveal melanoma (UM) (Fig. 7B-H). Furthermore, T-SNE diagrams were employed to display the expression profiles of ZBTB11 at the single-cell level in AML, BRCA, glioma, HGG, RB, RCC, and UM (Fig. 7I-O).
3.8. Enrichment analyses of ZBTB11 Co-expressed Genes in LIHC
To investigate the biological functions and pathways of ZBTB11 in LIHC, ZBTB11 was found to be positively correlated with 12,209 genes, and negatively correlated with 7713 genes (Fig. 8A). Heatmaps of the top 50 genes associated with ZBTB11 are presented in Fig. 8B and Fig. 8C. Analyzing KEGG data, the co-expressed genes were enriched in JAK-STAT signaling pathway, ribosome, oxidative phosphorylation and proteasome (Fig. 8D). Biological processes exploration unveiled concentration in demethylation, translational elongation, NADH dehydrogenase complex assembly, and mitochondrial gene expression (Fig. 8E). Examination of cellular components highlighted clustering in basal part of the cell, site of DNA damage, PML body, mitochondrial protein complex, respiratory chain, and mitochondrial inner membrane (Fig. 8F). Additionally, molecular function analysis revealed significant enrichment for histone binding, helicase activity, activating transcription factor binding, structural constituent of ribosome, and oxidoreductase activity, acting on NAD(P)H (Fig. 8G).
3.9. lncRNA-miRNA-ZBTB11 network construction in LIHC
In recent years, numerous studies have demonstrated the significant role of long non-coding RNAs (lncRNAs) in tumorigenesis through the regulation of downstream mRNA expression by sequestering their target miRNAs [45]. As a result, we conducted an investigation of the lncRNA-miRNA network that potentially regulates ZBTB11 expression in LIHC. First, we screened the miRWalk, miRDB, and Targetscan 8.0 databases and identified 80 miRNAs that are potential targets of ZBTB11 mRNAs (Fig. 9A). Subsequently, K-M survival curve analysis revealed 22 miRNAs associated with the prognosis of LIHC patients (Fig. S8). However, only 12 target miRNAs could be retrieved from StarBase to predict their lncRNAs (Fig. 9B). Using cytoscape, we constructed the lncRNA-miRNA-ZBTB11 regulatory network for potential drug discovery targeting ZBTB11 (Fig. 9C).
3.10. Pan-cancer sensitivity of ZBTB11-related drugs
Based on the GDSC drug sensitivity results, the top five drugs negatively associated with ZBTB11 expression were NPK76-II-72-1, PHA-793887, PIK-93, Methotrexate, and I-BET-762 (Fig. 10A). The CTPR dataset showed the correlation between ZBTB11 mRNA expression levels and drug sensitivity, the top five drugs that were negatively related to ZBTB11 expression were BRD-K01737880, GSK-J4, isoliquiritigenin, belinostat, and indisulam (Fig. 10B). The CTD was also used to establish a network of interactions between ZBTB11 chemicals (Fig. 10C).