Construction of a novel 3-gene diagnostic signature related to senescence in intervertebral disc degeneration.

BACKGROUND: Numerous studies have manifested that cellular senescence involves in the pathogenesis of intervertebral disc degeneration (IDD). Here, we constructed a novel senescence-related genes (SRGs) signature for IDD. METHODS: Three data sets were derived from Gene Expression Omnibus (GEO) database and 370 SRGs were collected from cellAge database. Key module genes related to senescence in IDD were screened using "WGCNA" package. The "limma" package was employed to filter differentially expressed genes (DEGs) between control and IDD groups, and candidate genes were identified by intersecting up-regulated DEGs and module genes. Hub genes were screened by randomForest and LASSO regression analysis. Diagnosis performance of hub genes was assessed and verified by receiver operating characteristic (ROC) curve. Diagnosis biomarkers of IDD were identified based on AUC > 0.7. Signaling pathway enrichment analysis of biomarkers was performed using "clusterProfiler" package. Immune cells infiltration was evaluated by "MCP-counter" and "GSVA" packages. RESULTS: A total of 625 module genes and 384 DEGs were obtained, then 28 candidate genes related to senescence in IDD were screened. By machine learning, 4 hub genes were identified with good diagnostic performance, which were highly expressed in IDD samples. Furthermore, 3 biomarkers (BID, KANK2, and SMIM3) were screened with AUC > 0.7 in external datasets. Three biomarkers were mainly involved in nuclear factor (NF)-kappa B, TNF, IL-17, and NOD-like receptor signaling pathway, etc. Besides, BID, KANK2, and SMIM3 exhibited positive association with most immune cell infiltration. CONCLUSIONS: We identified 3 diagnostic biomarkers related to senescence in IDD, hoping to improve the treatment of IDD.