VWCE mRNA expression levels in various types of cancer
Information in the TIMER database uncovered that mRNA expression of VWCE was fundamentally lower in breast cancer tissues when compared with normal tissues (p < 0.01). As shown in Figure 1A, compared with normal tissues, the VWCE transcription levels were significantly lower in BLCA (bladder urothelial carcinoma), BRCA (breast invasive carcinoma), CHOL (cholangiocarcinoma), HNSC (head and neck squamous cell carcinoma), KICH (kidney chromophobe) and PRAD (prostate adenocarcinoma). Additionally, to confirm the differences in VWCE expression in BRCA, VWCE expression was analyzed using the DriverDBv3 databases. Figure 1B also showed that VWCE mRNA expression was lower in breast tumors than normal tissues.
Associations between VWCE expression profiles and clinicopathological parameters in breast cancer patients
We investigated the relationship between VWCE expression and the clinical characteristics of breast cancer patients using bc-GenExMiner 4.4. There were remarkably differential expression levels of VWCE mRNA in ER status (ER- > ER+, p < 0.0001, Figure 2 A), PR status (PR- > PR+, p < 0.0001, Figure 2 B), HER2 status (HER2+ > HER2-, p = 0.0006, Figure 2 C). However, no significant expression difference of VWCE mRNA was found in age and nodal status (Figure 2 D E). In addition, we found that there was significant difference between VWCE expression and subtypes (HER2-E > basal-like, p < 0.0001; luminal A < basal-like, p < 0.0001; luminal A < HER2-E, p < 0.0001; luminal B < HER2-E, p < 0.0001) (Figure 2 F). These results suggest that VWCE expression may serve as a potential diagnostic indicator in breast cancer.
Biological interaction network of VWCE
To decide the biological interaction network of VWCE in breast cancer, we analyzed the functional protein association network by the STRING database. The co-expression analysis revealed that VWCE was co-expressed with bone morphogenetic protein 2 (BMP2), twisted gastrulation protein homolog 1 (TWSG1), bone morphogenetic protein receptor type-2 (BMPR2), bone morphogenetic protein receptor type-1A (BMPR1A), bone morphogenetic protein 7 (BMP7), Bone morphogenetic protein 6 (BMP6), growth/differentiation factor 5 (GDF5), thrombospondin type-1 domain-containing protein 1 (THSD1), zinc finger protein 114 (ZNF114) and ran-binding protein 3 (RANBP3), whose correlation scores were 0.758, 0.659, 0.654, 0.639, 0.628, 0.608, 0.594, 0.581, 0.546, and 0.535, respectively (Figure 3A), suggesting that they may be functional partners in breast cancer.
To examine the function of the identified genes, biological analyses (GO enrichment and KEGG pathway analysis) were performed via the Enrichr online database (Figure 3B, Supplementary table). GO analysis results demonstrated that the biological processes of these proteins were mainly involved in positive regulation of pathway-restricted SMAD protein phosphorylation (GO: 0010862), positive regulation of bone mineralization (GO: 0030501), and regulation of pathway-restricted SMAD protein phosphorylation (GO: 0060393). Molecular function (MF) were mainly enhanced in BMP receptor binding (GO: 0070700), BMP receptor activity (GO: 0098821), and transmembrane receptor protein serine/threonine kinase binding (GO: 0070696). Cell component was significantly enriched in the spanning component of the plasma membrane (GO: 0044214), HFE-transferrin receptor complex (GO: 1990712), and spanning component of membrane (GO: 0089717). Therefore, we speculate that VWCE may be related to the occurrence and development of cancer.
Gene Set Cancer Analysis
To further understand the SNM, CNV, and pathway activity of these proteins, we performed the analysis by GSCALite (Figure 4). From the SNV module, the SNV frequency of BMPR2, VWCE, BMP2, RANBP3, and ZNF114 are in the top five, which are 19 %, 19 %, 15 %, 15 %, and 15 %, respectively. Among them, the variant types of VWCE, RANBP3, and ZNF114 are missense mutations and frame-shift-Del. On the CNV module, the main copy number variants of these genes include heterozygous amplification and heterozygous deletion. Thus, these data suggested that VWCE mutations were associated with its expression. Moreover, the results of pathway analyses revealed that the expression of VWCE was correlated with the activation or inhibition of multiple oncogenic pathways. VWCE expression mainly inhibits the apoptosis pathway, and the cell cycle pathway could be inhibited by THSD1 and BMP6 expression. However, the EMT pathway mainly is activated by TWSG1, THSD1, and GDF5 expression. Therefore, the biological interaction network of VWCE is engaged with protein complex formation, protein regulation, and cancer processes.
VWCE expression impacts the prognosis and correlates with immune infiltration level in breast cancer
To better understand the relevance and underlying mechanisms of VWCE expression in breast cancer, we investigated the relationship between the expression of VWCE and clinical characteristics of breast cancer patients in the Kaplan-Meier plotter databases. As shown in Figure 5 A, the high expression of VWCE was associated with better prognosis (HR = 0.67, p = 0.015). These results suggest that the expression level of VWCE can impact the prognosis of breast cancer. Moreover, we analyzed the correlation between somatic copy number alterations and the abundance of immune infiltrates of VWCE (Figure 5 B). SCNA module provides the comparison of tumor infiltration levels among tumors with different somatic copy number alterations for a given gene. SCNAs are defined by GISTIC 2.0, including deep deletion (-2), arm-level deletion (-1), diploid/normal (0), arm-level gain (1), and high amplification (2). Box plots are presented to show the distributions of each immune subset at each copy number status in breast cancer.
Previous studies reported that the survival time of patients in several cancers depends on the number and activity of tumor-infiltrating lymphocytes [26, 27]. Therefore, we investigated whether VWCE expression was correlated with immune infiltration levels in breast cancer. We assessed the correlations of VWCE expression with immune infiltration levels in breast cancer using the TIMER database. The results showed that VWCE expression significantly correlated with tumor purity and significant correlations with CD4+ T cell, macrophages, and dendritic cell infiltration levels in breast cancer (Figure 5 C). However, the clinical impact of immune cells in breast cancer remains poorly understood. Hence, there is a great need for a more comprehensive analysis of tumor immunity to better understand. Next, we analyzed immune cells (B cell, CD8+ T cell, CD4+ T cell, macrophages, neutrophils, and dendritic cells) on the prognosis in breast cancer, the expression of these immune cells were divided into high and low levels by using the median expression. The results showed that high levels of B cell, CD8+ T cell, CD4+ T cell, neutrophils, and dendritic cell (Figure 5 D) were significantly associated with better survival (p < 0.05), whereas macrophages were not (p > 0.05), it is worth further research and exploration. These findings strongly indicated that VWCE played an important role in immune infiltration in breast cancer.
Regulation of immune molecules by VWCE
To further study the regulation of immune molecules by VWCE in breast cancer, we conducted an integrated analysis to predict correlations between VWCE expression and lymphocytes and immunomodulators using the TISIDB database (Figure 6). Interestingly, we found the greatest correlation between VWCE expression and TILs included mast (Spearman: rho = 0.362, p < 2.2e-16), macrophages (Spearman: rho = 0.327, p < 2.2e-16), Tfh (Spearman: rho = 0.325, p < 2.2e-16), and Th1 (Spearman: rho = 0.323, p < 2.2e-16) (Figure 6 A). Immunomodulators can be further classified into immunoinhibitors, immunostimulators, and major histocompatibility complex (MHC) molecules[28]. The correlations between VWCE expression levels and immunoinhibitors were showed in Figure 6 B, among these immunoinhibitors, VWCE expression had the strongest correlation with TGFB1 (Spearman: rho = 0.34, p < 2.2e-16), CD244 (Spearman: rho = 0.261, p = 2.5e-18), PDCD1 (Spearman: rho = 0.252, p = 2.5e-17), and LGALS9 (Spearman: rho = 0.238, p = 1.5e-15). Figure 6 C showed correlations between VWCE expression and immunostimulators, and the immunostimulators displaying the greatest correlations included C10orf54 (Spearman: rho = 0.437, p < 2.2e-16), TNFRSF8 (Spearman: rho = 0.375, p < 2.2e-16), TNFRSF4 (Spearman: rho = 0.374, p < 2.2-16), and TNFRSF25 (Spearman: rho = 0.366, p < 2.2e-16). Next, we compared correlations between VWCE expression and MHC molecules (Figure 6 D), and the MHC molecules displaying the greatest correlations included HLA-DPB1 (Spearman: rho = 0.302, p < 2.2e-16), HLA-E (Spearman: rho = 0.299, p < 1.26e-24), HLA-DRB1 (Spearman: rho = 0.266, p < 3.62e-19), and HLA-DMA (Spearman: rho = 0.259, p < 3.55e-18). Therefore, VWCE may be involved regulating the above immune molecules.
Correlation analysis between VWCE expression and immune marker sets
To detect the relationship between VWCE and the diverse immune infiltrating cells, we focused on the correlations between VWCE and immune marker sets of various immune cells in BRCA using the TIMER database. The immune cells analyzed in BRCA tissues included CD8+ T cell, T helper 1 (Th1), follicular helper T (Tfh), M1 Macrophage, and M2 Macrophage (Table 1). VWCE expression level was significantly correlated with CD8+ T cell, Th1, M1 macrophage, and M2 macrophage. The strongest correlations were with immune cell markers for CD8+ T cell (CD8B), Th1 (T-bet, STAT1), M1 macrophage (COX2), and M2 macrophage (MS4A4A). Next, we further investigated the correlation between VWCE and the diverse immune infiltrating cells using GEPIA database (Figure 7, Table 2). Specifically, VWCE expression showed significant correlation with the expression of markers of Th1, STAT4 (r = 0.11, p = 0.00021), STAT1 (r = -0.12, p = 6e-05), and M2 macrophage markers, CD163 (r = 0.12, p = 2.5e-05), VSIG4 (r = 0.22, p = 3.3e-14), and MS4A4A (r = 0.28, p = 0), but not with Tfh and M1 macrophage. Thus, these results suggested that VWCE may be involved in Th1 and M2 macrophage infiltrates.
We further analyzed the correlation between VWCE expression and these markers by immunohistochemistry (Figure 8). The levels of the expression were quantitated by scoring staining intensity, including negative (-) and weak (+) staining, moderate (++) and strong (+++) staining, respectively [29]. We found that VWCE mainly localized in the extracellular. VWCE showed strong expression in normal tissues (Figure 8 A), but low expression in breast cancer tissues (Figure 8 B). The expression level of STAT1 was high in breast cancer tissues (Figure 8 C), while the expression level of MS4A4A was relatively low in breast cancer tissues (Figure 8 D). The results further revealed that STAT1 was negative correlation with the expression of VWCE, but MS4A4A was positive correlation with the expression of VWCE. Taken together, the expression of VWCE relates to infiltration levels of Th1 and M2 macrophages.