ABCD3 is a prognostic biomarker for glioma and associated with immune infiltration: A study based on oncolysis of gliomas

Abstract

Background: Gliomas are the most lethal primary brain tumors and are still a major therapeutic challenge. Oncolytic virus therapy is a novel and effective means for glioma. However, little is known about gene expression changes during this process and their biological functions on glioma clinical characteristics and immunity.

Methods: The RNA-seq data after oncolytic virus EV-A71 infection on glioma cells were analyzed to screen significantly downregulated genes. Once ABCD3 was selected, The Cancer Genome Atlas (TCGA), Chinese Glioma Genome Atlas (CGGA), Genotype-Tissue Expression (GTEx), Clinical Proteomic Tumor Analysis Consortium (CPTAC), and Human Protein Atlas (HPA) data were used to analyze the relationship between ABCD3 expression and clinical characteristics in glioma. We also evaluated the influence of ABCD3 on the survival of glioma patients. CIBERSORT and Tumor Immune Estimation Resource (TIMER) were also used to investigate the correlation between ABCD3 and cancer immune infiltrates. Gene set enrichment analysis (GSEA) was performed to functionally annotate the potential functions or signaling pathways related to ABCD3 expression.

Results: ABCD3 was among the top 5 downregulated genes in glioma cells after oncolytic virus EV-A71 infection and was significantly enriched in several GO categories. Both the mRNA and protein expression levels of ABCD3 were upregulated in glioma samples and associated with the prognosis and grades of glioma patients. The Kaplan-Meier (K-M) curve analysis revealed that patients with high ABCD3 expression had shorter disease-specific survival (DSS) and overall survival (OS) than those with low ABCD3 expression. Moreover, ABCD3 expression could affect the immune infiltration levels and diverse immune marker sets in glioma. A positive correlation was found between ABCD3 and macrophages and active dendritic cells in the microenvironment of both the GBM and LGG. Gene sets including the plk1 pathway, tyrobp causal network, ir-damage and cellular response, and interleukin-10 signaling showed significant differential enrichment in the high ABCD3 expression phenotype.

Conclusion: Our results suggested that ABCD3 could be a potential biomarker for glioma prognosis and immunotherapy response and also further enriched the theoretical and molecular mechanisms of oncolytic virus treatment for malignant gliomas.

Keywords: ABCD3; biomarker; glioma; immune infiltrates; oncolytic virus.

Figure 1

Gene Ontology (GO) analysis for all downregulated genes after EV-A71 infection on glioma cells. The distribution of downregulated genes from EV-A71 groups was obtained relative to the mock infection. GO analysis for all downregulated genes was performed and many enriched GO categories were found.

Figure 2

The mRNA and protein expression levels of ABCD3 are upregulated in glioma samples. (A) Differential expression of ABCD3 in both GBM and LGG compared with matched control samples. (B) Differential expression of ABCD3 in different glioma grades. (C) Differential ABCD3 protein expression levels in glioblastoma compared with normal tissues based on the data from the CPTAC. (D) Immunohistochemistry images of ABCD3 protein expression in glioma tissues and their normal controls. ns, no significance. *, P < 0.05; **, P < 0.01.

Figure 3

ABCD3 expression is associated with the clinical characteristics of glioma patients. (A) ABCD3 expression is lower in the 1p/19q co-deletion (Codel) groups than in the no co-deletion groups in all three WHO grade gliomas. (B–E) Overall survival and disease-specific survival curve of GBM and LGG patients based on differential ABCD3 expression. (F) Univariate Cox analysis of ABCD3 expression and other clinical pathological factors for GBM and LGG.

Figure 4

Relationship between ABCD3 expression and tumor-infiltrating immune cells. The proportion of 24 subpopulations of immune cell infiltration between the high and low expression of ABCD3 in GBM (A) or LGG (B) samples. Macrophages were mostly increased in the high ABCD3 expression group compared with the low ABCD3 expression group, while pDC cells were apparently decreased in the high expression group in both GBM and LGG. The macrophage infiltrate correlated with both GBM (C) and LGG (D). ns, no significance; *, P < 0.05; ***, P < 0.001.

Figure 5

The ABCD3 expression level has a correlation with the tumor-infiltrating levels from TIMER. (A) ABCD3 expression has positive correlations with infiltrating levels of B cell, CD8⁺ T cell, macrophage, and neutrophil, while it has a negative correlation with dendritic cells in GBM. (B) ABCD3 expression has a significant positive correlation with all immune infiltration cells in LGG.

Figure 6

GSEA analysis of LGG or GBM samples with high or low ABCD3 expression phenotype. (A) GSEA results showing the top 5 enriched gene sets in the LGG samples with high ABCD3 expression based on NES. (B) GSEA results showing the top 5 enriched gene sets in the LGG samples with low ABCD3 expression based on NES. (C) GSEA results showing the top 5 enriched gene sets in the GBM samples with high ABCD3 expression based on NES. (D) GSEA results showing the top 5 enriched gene sets in the GBM samples with low ABCD3 expression based on NES.