The pro-invasive factor COL6A2 serves as a novel prognostic marker of glioma

BACKGROUND: Glioma is an incurable malignant lesion with poor outcome characterized by easy recurrence after surgery with or without radiotherapy and chemotherapy. Studies have shown that COL6A2 is closely related to the tumorigenesis and development of a variety of tumors. However, the role of COL6A2 in glioma and the relationship between COL6A2 and tumor infiltrating immune cells remain unclear. METHODS: Western blot, real-time PCR, a tissue microarray and immunohistochemistry were applied to detect COL6A2 mRNA and protein amounts in glioma, and all experiments were repeated three times. A tissue microarray of glioma samples was used for prognostic analysis. Detection of COL6A2 co-expression with immune genes using immunohistochemical methods, and tumor modeling using nude mice for prevention and treatment studies. Based on the mRNA expression of COL6A2, patients with glioma in TCGA were divided into the low and high COL6A2 expression groups, and GO and KEGG pathway analyses were performed. A PPI network was constructed using STRING, and the associations of COL6A2 with tumor-infiltrating immune cells and immune genes were analyzed in the CIBERSORT and TISIDB databases. COL6A2 mRNA and protein amounts were increased in glioma. RESULTS: Multiple-database and tissue microarray analyses showed that COL6A2 expression in glioma was associated with poor prognosis, Tissue microarray showed that COL6A2 was the highest expressed in WHO IV and significantly higher in TCGA-GBM than in TCGA-LGG. Immunohistochemistry can well demonstrate the co-expression of COL6A2 with immune genes in a tumor model established in nude mice, showing that interference with COL6A2 expression may have an inhibitory effect on tumors. The mRNA expression of COL6A2 was involved in 22 KEGG pathways, and GSEA analysis showed that 28 and 57 gene sets were significantly enriched at nominal p values <0.01 and <0.05, respectively, protein network revealed a tight interaction between COL6A2 and SPARC. The CIBERSORT database indicated that COL6A2 was correlated with 15 types of tumor-infiltrating immune cells, including M2 macrophages, CD8 T cells, neutrophils, gamma delta T cells, activated CD4 memory T cells, follicular helper T cells, M0 macrophages, M1 macrophages, regulatory T cells (Tregs), activated NK cells, eosinophils, activated mast cells, monocytes, activated dendritic cells, and resting CD4 memory T cells. The TISIDB database indicated that COL6A2 was significantly correlated with lymphocytes such as regulatory T cell, Type 17 T helper cell, Type 1 T helper cell, and immunomodulatory genes. In addition, COL6A2-related immune regulatory genes show that most immune regulatorygenes have prognostic value for glioma, and high-risk immune genes are notconducive to the survival of glioma patients. CONCLUSIONS: COL6A2-related immune regulatory genes show that most immune regulatory genes have prognostic value for glioma, and high-risk immune genes are not conducive to the survival of glioma patients. COL6A2 may be a novel potential prognostic biomarker of glioma and associated with tumor-infiltrating immune cells in the tumor microenvironment, and interference with COL6A2 expression can inhibit tumor growth, which suggests COL6A2 as a potential target for future treatment.