Integrated Bioinformatic Analysis of the Shared Molecular Mechanisms Between Osteoporosis and Aortic Stenosis

Introduction: Osteoporosis (OP) and aortic stenosis (AS) are highly prevalent age-related disorders that frequently coexist. Epidemiological studies suggest a pathological link between OP and AS beyond age, yet the molecular mechanisms underlying this bone-vascular axis remain poorly defined. This study aimed to identify shared genes and pathways contributing to the comorbidity of OP and AS. Methods: Publicly available AS and OP transcriptomic datasets were retrieved from the GEO database. Weighted gene co-expression network analysis (WGCNA) and differential gene expression (DEG) analysis were conducted to identify disease-associated genes. Candidate hub genes were screened through protein-protein interaction (PPI) network analysis using twelve network topology algorithms. High-confidence genes were obtained by intersecting candidates with AS-related genes from the Comparative Toxicogenomics Database (CTD). Independent cohorts were used to validate candidate genes, and least absolute shrinkage and selection operator (LASSO) regression was performed to assess their diagnostic potential. Results: WGCNA revealed 665 shared genes enriched in immune and inflammatory processes, cell adhesion, and glycosaminoglycan biosynthesis. PPI network analysis identified 32 candidate hub genes, and integration with CTD yielded 15 high-confidence genes. Validation across independent datasets confirmed dysregulated expression of CD4, GZMB, and SDC1 in both AS and OP samples. ROC analysis demonstrated high diagnostic accuracy of these genes, with a combined AUC of 0.94. Discussion: These findings highlight immune and inflammatory pathways as convergent mechanisms driving both AS and OP. The hub genes CD4, GZMB, and SDC1 participate in immune regulation and extracellular matrix remodeling, suggesting their involvement in the shared pathogenesis of skeletal and cardiovascular degeneration. result: In total, 6 and 5 co-expression modules were found in the AS dataset and OP dataset, respectively, by WGCNA. WGCNA identified 665 common genes enriched in processes such as tertiary granule, specific granule, cell adhesion molecules, tuberculosis, and glyosaminoglyan biosynthesis. Of these, 13 were screened as candidate genes by 12 prediction algorithms in Cytoscape. Another analysis identified 161 common DEGs enriched in extracellular matrix organization, immunological synapse, extracellular matrix structural constituent conferring tensile strength, and natural killer cell mediated cytotoxicity. Of these, 19 were recognized as candidate genes by Cytoscape in the same way. 15 hub genes were selected by intersecting candidate genes with genes downloaded from CTD as hub genes, including CD226, CD247, CD38, CD4, CD96, FCGR2B, GNG2, GRB2, GAMB, HSD17B6, ITGB2, KSD17B6, ITGB2, KLRB1, NGF, PECAM1, SDC1. The transcription factors (TFs)-hub genes interaction network was constructed, and 9 TFs were identified using TRRUST. These hub genes were validated another AS dataset and OP dataset. At the end, we find CD4, GZMB, SDC1 are upregulated remarkably in AS dataset, whereas SDC1 and ITGB2 are downregulated and GZMB is upregulated in OP dataset. Conclusion: Integrative bioinformatics identified CD4, GZMB, and SDC1 as key genes linking OP and AS, providing potential biomarkers and therapeutic targets for managing these age-related comorbidities.