Identification and validation of a risk model and molecular subtypes based on tryptophan metabolism-related genes to predict the clinical prognosis and tumor immune microenvironment in lower-grade glioma

Construction and validation of a risk model based on the key SNARE proteins to predict the prognosis and immune microenvironment of gliomas

Background

Synaptic transmission between neurons and glioma cells can promote glioma progression. The soluble N-ethylmaleimide-sensitive fusion factor attachment protein receptors (SNARE) play a key role in synaptic functions. We aimed to construct and validate a novel model based on the SNARE proteins to predict the prognosis and immune microenvironment of glioma.

Methods

Differential expression analysis and COX regression analysis were used to identify key SRGs in glioma datasets, and we constructed a prognostic risk model based on the key SRGs. The prognostic value and predictive performance of the model were assessed in The Cancer Genome Atlas (TCGA) and Chinese glioma Genome Atlas (CGGA) datasets. Functional enrichment analysis and immune-related evaluation were employed to reveal the association of risk scores with tumor progression and microenvironment. A prognostic nomogram containing the risk score was established and assessed by calibration curves and time-dependent receiver operating characteristic curves. We verified the changes of the key SRGs in glioma specimens and cells by real-time quantitative PCR and Western blot analyses.

Results

Vesicle-associated membrane protein 2 (VAMP2) and vesicle-associated membrane protein 5 (VAMP5) were identified as two SRGs affecting the prognoses of glioma patients. High-risk patients characterized by higher VAMP5 and lower VAMP2 expression had a worse prognosis. Higher risk scores were associated with older age, higher tumor grades, IDH wild-type, and 1p19q non-codeletion. The SRGs risk model showed an excellent predictive performance in predicting the prognosis in TCGA and CGGA datasets. Differentially expressed genes between low- and high-risk groups were mainly enriched in the pathways related to immune infiltration, tumor metastasis, and neuronal activity. Immune score, stromal score, estimate score, tumor mutational burden, and expression of checkpoint genes were positively correlated with risk scores. The nomogram containing the risk score showed good performance in predicting the prognosis of glioma. Low VAMP2 and high VAMP5 were found in different grades of glioma specimens and cell lines.

Conclusion

We constructed and validated a novel risk model based on the expression of VAMP2 and VAMP5 by bioinformatics analysis and experimental confirmation. This model might be helpful for clinically predicting the prognosis and response to immunotherapy of glioma patients.