Integration of bioinformatics and multi-layered experimental validation reveals novel functions of acetylation-related genes in intervertebral disc degeneration

Abstract

Background

The molecular mechanisms underlying intervertebral disc degeneration (IDD) remain poorly understood. The purpose of this work is to elucidate key molecules and investigate the roles of acetylation-related RNAs and their associated pathways in IDD.

Method

Datasets GSE70362 and GSE124272 were obtained from the Gene Expression Omnibus (GEO) and combined to investigate differentially expressed genes (DEGs) associated with acetylation in IDD patients compared to healthy controls. Critical genes were pinpointed by integrating GO, KEGG and PPI networks. Furthermore, CIBERSORTx analysis was used to investigate the differences in immune cell infiltration between different groups and the biological processes (BP), cellular components (CC) and molecular functions (MF) were calculated by GSEA and GSVA. In addition, The single-cell database GSE165722 was incorporated to validate the specific expression patterns of hub genes in cells and identify distinct cell subtypes. This provides a theoretical basis for a more in-depth understanding of the roles played by critical cell subtypes in the process of IDD. Subsequently, tissues from IVD with varying degrees of degeneration were collected to corroborate the key DEGs using western blot, RT-qPCR, and immunofluorescence staining.

Results

By integrating various datasets and references, we identified a total of 1620 acetylation-related genes. These genes were subjected to a combined analysis with the DEGs from the databases included in this study, resulting in the discovery of 358 acetylation-related differentially expressed genes (ARDEGs). A comparative analysis with differentially expressed genes obtained from three databases yielded 19 ARDEGs. The PPI network highlighted the top 10 genes (IL1B, LAMP1, PPIA, SOD2, LAMP2, FBL, MBP, SELL, IRF1 and KHDRBS1) based on their protein interaction relationships. CIBERSORTx immune infiltration analysis revealed a moderate positive correlation between the gene IL1β and Mast.cells.activated, as well as a similar correlation between the gene IRF1 and Mast.cells.activated. Single-cell dataset was used to identify cell types and illustrate the distribution of hub genes in different cell types. The two cell types with the highest AUCell scores (Neutrophils and Monocytes) were further explored, leading to the subdivision of Neutrophils into two new cell subtypes: S100A9-type Neutrophils and MARCKS-type Neutrophils. Monocytes were labeled as HLA-DRA9-type Monocytes and IGHG3-type Monocytes. Finally, molecular biology techniques were employed to validate the expression of the top 10 hub genes. Among them, four genes (IL1β, SOD2, LAMP2, and IRF1) were confirmed at the gene level, while two (IL1β and SOD2) were validated at the protein level.

Conclusion

In this study, we carried out a thorough analysis across three databases to identify and compare ARDEGs between IDD patients and healthy individuals. Furthermore, we validated a subset of these genes using molecular biology techniques on clinical samples. The identification of these differently expressed genes has the potential to offer new insights for diagnosing and treating IDD.