Abstract
INTRODUCTION: Clinical studies reveal bidirectional links between sarcopenia (SP) and diabetic nephropathy (DN). Damage to mitochondria in DN may result in diminished energy production, which consequently triggers SP. Consequently, mitochondria seem to function as critical nodes connecting DN with SP. The objective of this research was to pinpoint biomarkers associated with mitochondrial dysfunction in DN and SP. METHODS: By analyzing the Gene Expression Omnibus (GEO) repository, we identified shared differentially expressed genes (DEGs) in the DN (GSE96804, GSE30528) and SP (GSE1428, GSE136344) datasets that displayed similar expression trends in mitochondrial genes. Using Least Absolute Shrinkage and Selection Operator (LASSO), Support Vector Machine (SVM), Extreme Gradient Boosting (XGB), and Random Forest (RF) algorithms, we identified three key mitochondrial hub genes. Diagnostic nomograms were created to predict DN and SP risk. We assessed immune infiltration using CIBERSORT and built a drug-gene network with Cytoscape. Molecular docking determined binding affinities between potential drugs and hub genes, which were validated in the datasets. RESULTS: Analysis of GEO datasets identified 80 shared DEGs between DN and SP, including 10 mitochondria-related genes. Utilizing four machine learning algorithms (LASSO, SVM, XGBoost, RF), we pinpointed three mitochondrial hub genes. Subsequent validation confirmed two key mitochondria-related genes - HTT and TTC19 - as shared diagnostic biomarkers for both DN and SP. These biomarkers demonstrated strong diagnostic power (AUC >0.8), leading to the construction of diagnostic nomograms. Immune infiltration analysis revealed elevated M1 macrophages in DN and increased M2 macrophages in SP, with both biomarkers showing significant correlations with various immune cells. Gene set enrichment analysis linked HTT and TTC19 to mitochondrial metabolic processes. Crucially, in silico drug prediction identified 156 potential drugs, and molecular docking confirmed a high binding affinity between acetaminophen and the HTT protein. CONCLUSION: Our study identifies HTT and TTC19 as novel mitochondria-immune related biomarkers common to both DN and SP, providing insights into their shared pathogenesis involving mitochondrial dysfunction and immune dysregulation. The computational prediction of acetaminophen interaction highlights a potential avenue for therapeutic exploration. Further clinical and mechanistic studies are warranted to validate these findings and elucidate the underlying pathways.