Identification and analysis of cellular senescence-associated signatures in diabetic kidney disease by integrated bioinformatics analysis and machine learning

Front Endocrinol (Lausanne). 2023 Jun 16:14:1193228. doi: 10.3389/fendo.2023.1193228. eCollection 2023.

Abstract

Background: Diabetic kidney disease (DKD) is a common complication of diabetes that is clinically characterized by progressive albuminuria due to glomerular destruction. The etiology of DKD is multifactorial, and numerous studies have demonstrated that cellular senescence plays a significant role in its pathogenesis, but the specific mechanism requires further investigation.

Methods: This study utilized 5 datasets comprising 144 renal samples from the Gene Expression Omnibus (GEO) database. We obtained cellular senescence-related pathways from the Molecular Signatures Database and evaluated the activity of senescence pathways in DKD patients using the Gene Set Enrichment Analysis (GSEA) algorithm. Furthermore, we identified module genes related to cellular senescence pathways through Weighted Gene Co-Expression Network Analysis (WGCNA) algorithm and used machine learning algorithms to screen for hub genes related to senescence. Subsequently, we constructed a cellular senescence-related signature (SRS) risk score based on hub genes using the Least Absolute Shrinkage and Selection Operator (LASSO), and verified mRNA levels of hub genes by RT-PCR in vivo. Finally, we validated the relationship between the SRS risk score and kidney function, as well as their association with mitochondrial function and immune infiltration.

Results: The activity of cellular senescence-related pathways was found to be elevated among DKD patients. Based on 5 hub genes (LIMA1, ZFP36, FOS, IGFBP6, CKB), a cellular senescence-related signature (SRS) was constructed and validated as a risk factor for renal function decline in DKD patients. Notably, patients with high SRS risk scores exhibited extensive inhibition of mitochondrial pathways and upregulation of immune cell infiltration.

Conclusion: Collectively, our findings demonstrated that cellular senescence is involved in the process of DKD, providing a novel strategy for treating DKD.

Keywords: bioinformatics analysis; cellular senescence; diabetic kidney disease; immune cell infiltration; machine learning; mitochondrial function.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Cellular Senescence / genetics
  • Computational Biology
  • Cytoskeletal Proteins
  • Diabetes Mellitus*
  • Diabetic Nephropathies* / genetics
  • Humans
  • Kidney
  • Machine Learning

Substances

  • LIMA1 protein, human
  • Cytoskeletal Proteins

Grants and funding

This work was supported by grants from the National Natural Science Foundation of China (NSFC 82270872) and the Natural Science Foundation of Guangdong (2022A1515012249).