Aberrant DNA polymerase beta expression is associated with dysregulated tumor immune microenvironment and its prognostic value in gastric cancer

Abstract

Background: Gastric cancer is caused by different exogenous risk factors. Polymerase beta (POLB) is critical to repair oxidative and alkylating-induced DNA damage in genome maintenance. It is unknown whether overexpression of POLB genes in GC modulates tumor immunogenicity and plays a role in its prognostic value.

Methods: RNA-Seq of GC data retrieved from TCGA and GEO database and patient survival were compared using Kaplan-Meier statistical test. The TIMER algorithm was used to calculate the abundance of tumor-infiltrating immune cells. Furthermore, ROC analysis was applied to evaluate the prognostic value of POLB overexpression.

Results: Our data analysis of TCGA and GEO gastric cancer genomics datasets reveals that POLB overexpression is significantly associated with intestinal subtypes of stomach cancer. In addition, POLB overexpression is associated with low expression of innate immune signaling genes. In contrast, POLB-overexpressed tumor harbors high mutation frequency and MSI score. Furthermore, POLB-overexpressed tumor with high immune score exhibits a better prognosis. Interestingly, our ROC analysis results suggested that POLB overexpression has a potential for prognostic markers for stomach cancer.

Conclusions: Our analysis suggests that aberrant POLB overexpression in stomach cancer impacts the diverse aspects of tumor immune microenvironment. In addition, POLB might be a potential prognosis marker and/or an attractive target for immune-based therapy in GC. However, our observation still requires further experimental-based scientific validation studies.

Fig. 1

POLB is overexpressed in gastric tumor and modulates in subtypes of stomach cancer. A Percent of POLB overexpression in stomach cancer (32%); B expression of POLB is high in tumor (T) versus normal non-cancerous tissues; C Stomach subtypes including intestinal, diffuse, tubular, singlet, mucin stomach adenocarcinoma; D Molecular subtypes of stomach cancer in POLB high- versus low-expressed tumor with chromosomal instability (CIN), microsatellite instability (MSI), Epstein-Barr virus infected (EBV), DNA polymerase E (POLE), genetically stable (GS); E Clinical characteristics of the intestinal and diffuse subtypes of stomach cancer from TCGA (lower panel) and GSE13861(upper panel). The Mann-Whitney U test analysis was performed using graph pad prism. * represent *P < 0.05; ** P < 0.01; ***P < 0.001; ****P < 0.0001 and ns represent no statistical significance difference

Fig. 2

The landscape of the tumor immune microenvironment and its  impact on gastric cancer. (A) Immune, stromal (B) and ESTIMATE score (C) of the tumor with high versus low POLB mRNA expression; (D) immune score of DTSA tumor with POLB high versus low; (E) immune score of ITSA with POLB high versus low POLB expression; (F) estimate score of DTSA versus ITSA; (G) overexpression of POLB is associated with low level of infiltration of immune cell to the tumor microenvironment. Correlation of immune infiltration versus the POLB expression in tumor. The correlation of POLB expression versus immune cell (CD8 + T, CD4 + T, B cells macrophage and DC cells); the Mann-Whitney U test was used to compare the expression of POLB versus immune score as well as stromal score. Spearman analysis was performed the correlation coefficient of POLB and immune infiltration. Rho represent correlation coefficient; * represent *P < 0.05; ****P < 0.0001 and ns represent no statistical significance difference

Fig. 3

Heatmap of innate immune signaling gene expression versus POLB mRNA expression in stomach cancer. A Heatmap of innate immune signaling genes (IFNB, ISG15, ISG20, CCL5, CXCL10, IRF1, IRF3, IRF7, STING) expression versus POLB mRNA expression in stomach cancer from TCGA database; B Heatmap of POLB mRNA expression versus innate immune signaling genes from GSE13861 database; C The relationship between the POLB expression versus immune check point genes (CTLA4, CD274 PDCD1, PDCD1GL2) from TCGA database; D POLB expression versus immune check point genes (CTLA4, CD274 PDCD1, PDCD1GL2) from GSE13861database. The Spearman’s correlation was performed to analyze using GraphPad. Rho represent the correlation coefficient

Fig. 4

The co-occurrence of POLB expression with high immune score is associated with better GC patient overall survival. A Kaplan–Meier survival analysis of POLB expression (low versus high) groups with high immune score; B The overall survival of diffuse stomach cancer subtypes in patients with tumor harbor high immune score with high POLB high expression versus low; C The overall survival of intestinal stomach cancer subtypes in patients with tumor harbor high immune score with POLB high expression versus low; D The overall survival of patients with high CTLA4 and high POLB expression versus low groups; E The overall survival of patients with high PD-1 with high POLB expression versus low groups. For Kaplan–Meier curves, P-values, were generated by log-rank tests. p < 0.05 was considered as statistically significant

Fig. 5

Diagnostic ROC curves of DNA POLB gene expression. A Diagnostic ROC curves of POLB expression from TCGA database; B Diagnostic ROC curves of POLB gene expression from GSE13861 database; C Diagnostic ROC values of TCGA and GSE13861 database with subtypes of stomach cancer