Comprehensive investigation in oncogenic functions and immunological roles of NCBP2 and its validation in prostate cancer

Abstract

Background: Nuclear cap-binding protein 2 (NCBP2), as the component of the cap-binding complex, participates in a number of biological processes, including pre-mRNA splicing, transcript export, translation regulation and other gene expression steps. However, the role of NCBP2 on the tumor cells and immune microenvironment remains unclear. To systematically analyze and validate functions of NCBP2, we performed a pan-cancer analysis using multiple approaches.

Methods: The data in this study were derived from sequencing, mutation, and methylation data in the TCGA cohort, normal sample sequencing data in the GTEx project, and cell line expression profile data in the CCLE database.

Results: Survival analyses including the Cox proportional-hazards model and log-rank test revealed the poor prognostic role of NCBP2 in multiple tumors. We further validated the oncogenic ability of NCBP2 in prostate cancer cell lines, organoids and tumor-bearing mice. A negative correlation was observed between NCBP2 expression and immune score by the ESTIMATE algorithm. Simultaneously, the NCBP2-induced immunosuppressive microenvironment might be related to the decline in CD8⁺T cells and the increase in regulatory T cells and neutrophils, examined by flow cytometry experiments for NCBP2 overexpressed tumor-bearing mice.

Conclusion: This research offered strong proof supporting NCBP2 as the prognostic marker and the therapeutic target in the future.

Keywords: NCBP2; Pan-cancer; Prognosis; Prostate cancer; Therapeutic target; Tumor microenvironment.

Fig. 1

Overview of NCBP2 expression in pan-cancer. (A) Differential expression analysis of NCBP2 between 33 tumor types and normal tissues. (B) Mean expression of NCBP2 in TCGA. (C) Mean expression of NCBP2 in GTEx. (D) Mean expression of NCBP2 in CCLE. The yellow lines represent the maximum values for the Y-axis. The red dots represent the expression of NCBP2 in each tumor, normal tissue, and cell line. The blue lines represent the average expression of NCBP2 in TCGA, GTEx, and CCLE. The grey lines represent the position of each tumor, normal tissue, and cell line on the X-axis. (E) The comparison of NCBP2 expression between adjacent normal tissues and PCa. (F–J) The distribution of NCBP2 expression in TCGA-PRAD classified by T, N, M stage, Gleason score and PSA. (K) Molecular interaction network of NCBP2. Representative immunohistochemical staining images of NCBP2 in (L) adjacent normal and (M) PCa tissues. Scale bars were 50 μm. (N) The comparison of NCBP2 H-scores between adjacent normal (n = 10) and PCa tissues (n = 10). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig. 2

Gene mutation and DNA methylation profiles of NCBP2. (A) CNV alteration frequency. (B) Correlations between NCBP2 expression and methylation. (C, D) Associations of NCBP2 expression levels with copy-number values and promoter methylation in PRAD. Top 10 mutant genes in the (E) NCBP2 low-expression and (F) high-expression groups. Mutant genes were ordered by mutation frequency. (G) Comparison of gene mutations between low-NCBP2 and high-NCBP2 groups in PRAD. *p < 0.05; **p < 0.01; ***p < 0.001.

Fig. 3

Prognostic capability of NCBP2. (A–D) Forest plots demonstrating survival analysis of NCBP2 on OS, DFI, DSS and PFI. Specific tumors (p < 0.05, HR> 1) were highlighted in red.

Fig. 4

Potential biological functions of NCBP2 in PRAD. (A–C) Summary of GSEA results for NCBP2 based on GO, KEGG, and Reactome databases. The value of x axis means the distribution of the fold change of gene expression transformed by log2 in each pathway. The height of the peak represents the frequency of distribution of genes in each pathway. The top 20 enriched pathways with the smallest P values were selected from the GO, KEGG, and Reactome databases, respectively. (D) GSVA for NCBP2 based on the Hallmark dataset in pan-cancer. (E) Relationships between NCBP2 expression and Hallmark pathways in PRAD. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig. 5

Experimental validation of NCBP2 in PCa. (A) Western blot validation of NCBP2 knockdown using shRNA in PC3 and DU145 cells. (B) The biological functions of NCBP2 on PRAD cell lines were verified by CCK-8 assays. (C) Transwell migration assays in NCBP2-knockdown PC3 and DU145 cells. Representative images in the left panels and statistical comparisons in the right panels. There were there replicates in each group. (D) Western blot validation of NCBP2 overexpression in the Pten^PC-/- cells after lentivirus infection. (E) Representative organoid images. Scale bars were 210um. (F-G) The comparison of organoid diameters and organoid formation efficiency. There were there replicates in each group. (H) Western blot validation of NCBP2 overexpression in the Pten^pc-/-; Trp53^pc-/-; Smad4^pc-/- -Luc cells after lentivirus infection. (I) Pten^pc-/-; Trp53^pc-/-; Smad4^pc-/- -Luc cells were injected into the prostate of C57BL/6 mice and bioluminescent images were taken on day 10. (J) Representative images of the PCa in situ. (K) The comparison of tumor volumes on week 3. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig. 6

Immunocorrelation analysis of NCBP2. (A–I) Correlations between NCBP2 expression and immune activation, immune suppressive, MHC, chemokine, chemokine receptor genes, tumor purity, stromal score, estimate score, immune score, immune checkpoints, TMB and MSI. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig. 7

Altered immune cell infiltrations. (A) Correlation of NCBP2 and immune cell infiltrations by the ImmuCellAI algorithm. (B–D) Flow cytometric analysis of CD45⁺CD3⁺CD8⁺T cells (B), CD45⁺Ly6G⁺CD11b⁺ Neutrophil (C) and CD45⁺CD3⁺CD8⁻Foxp3⁺ Treg cells in NC and NCBP2-OE mouse PCa (in situ injection of Pten^pc-/-; Trp53^pc-/-; Smad4^pc-/- cells). Representative plots (left panels) and statistical comparisons (right panels). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Fig. 8

Therapeutic response prediction. (A) Comparison of neoantigen burden between high and low NCBP2 expression groups. (B) Distribution of response to PD-L1 blockades in different groups. (C) Kaplan-Meier curves displaying prognostic differences between patients with high and low NCBP2 expression in the IMvigor210 cohort. (D–F) Correlation analysis between NCBP2 expression and the estimated IC50 of cisplatin, docetaxel and olaparib.