Identification of key genes and mechanistic exploration of metabolic reprogramming in osteoporosis
Yunfei He, Caihong Shi, Xin Yu, An WangObjective
To identify metabolism-related genes associated with osteoporosis and evaluate their diagnostic potential through integrative transcriptomic analysis and clinical validation.
Methods
Single-cell RNA sequencing data from GSE169396 were analyzed using the Seurat package to investigate interactions between monocytes and other cell types. Metabolic pathway alterations in osteoporosis were evaluated using single-sample gene set enrichment analysis. Differentially expressed genes were intersected with metabolism-related genes to identify differentially expressed metabolism-related genes. Key metabolic genes were screened using the least absolute shrinkage and selection operator, support vector machine-recursive feature elimination, and Boruta algorithms, and a logistic regression model was established to assess their diagnostic value. External quantitative reverse transcription polymerase chain reaction validation was performed using peripheral blood mononuclear cells from patients with osteoporosis and healthy controls. Receiver operating characteristic curve analysis was used to evaluate the predictive performance of individual genes and the combined model.
Results
Single-cell interaction analysis revealed close communication between monocyte subsets and hematopoietic stem cells, endothelial cells, cDC2, cDC1, T cells, and plasmacytoid dendritic cells. Significant alterations were observed in valine, leucine, and isoleucine biosynthesis; purine metabolism; fatty acid biosynthesis and elongation; ascorbate and aldarate metabolism; and steroid biosynthesis. Intersection analysis identified 11 differentially expressed metabolism-related genes from 232 metabolism-related genes and 395 differentially expressed genes. Integrative machine learning identified seven key metabolic genes: ADSL, BCAT1, BCAT2, PDE8A, NPR2, SOAT2, and XDH. External validation confirmed significant differential expression of these genes between patients with osteoporosis and healthy controls. The logistic regression model showed excellent diagnostic performance, with an area under the curve of 0.935, whereas all individual genes achieved area under the curve values above 0.7. Gene set enrichment analysis indicated that these genes were enriched in pathways related to G protein–coupled receptor activity.
Conclusions
We identified seven key metabolic genes associated with osteoporosis, namely, ADSL, BCAT1, BCAT2, PDE8A, NPR2, SOAT2, and XDH, and demonstrated their promising diagnostic potential. These findings provide new insights into the metabolic mechanisms underlying osteoporosis and may support the development of future diagnostic biomarkers and therapeutic targets.