Abstract
Background: Osteoporosis is a prevalent metabolic bone disorder with complex molecular underpinnings. Emerging evidence implicates endoplasmic reticulum stress (ERS) in its pathogenesis; however, systematic exploration of ERS-related genes (ERSRGs) remains limited. This study aimed to identify ERS-related differentially expressed genes (ERSRDEGs) in osteoporosis, construct a diagnostic model, and elucidate associated molecular mechanisms. Methods: Three osteoporosis datasets (GSE56815, GSE230665, and GSE7429) were integrated after batch effect correction and normalization. ERSRGs were curated from GeneCards, and ERSRDEGs were identified by intersecting co-differentially expressed genes (Co-DEGs) across datasets. Functional enrichment (gene set enrichment analysis [GSEA], gene set variation analysis [GSVA], Gene Ontology [GO], and Kyoto Encyclopedia of Genes and Genomes [KEGG]) and immune infiltration analyses were performed. Diagnostic models were developed using support vector machine (SVM) and least absolute shrinkage and selection operator (LASSO) regression, validated via receiver operating characteristic (ROC) curves, nomograms, and decision curve analysis. Experimental validation included immunohistochemistry and quantitative reverse transcription polymerase chain reaction (qRT-PCR) in ovariectomized (OVX) mice. Regulatory networks (TF-miRNA-RBP-drug) and protein structure predictions were generated using bioinformatic tools. Results: Fifty six ERSRDEGs were identified, enriched in apoptosis, autophagy, and cytokine signaling pathways. A diagnostic model comprising seven genes (CYB5R4, RAB1B, UFSP2, RNF13, SERP1, CES2, and C1QBP) demonstrated high accuracy (area under the curve (AUC) > 0.9) in both training and validation datasets. Immune infiltration analysis revealed distinct patterns of activated B cells, CD8(+) T cells, and macrophages between high- and low-risk groups. Regulatory networks highlighted interactions with 52 transcription factors (TFs), 42 miRNAs, and 27 therapeutic compounds. Experimental validation in OVX mice confirmed upregulated expression of C1QBP, CYB5R4, RAB1B, and UFSP2 at protein/mRNA levels, aligning with bioinformatic predictions. Conclusions: This study establishes ERSRDEGs as critical players in osteoporosis pathogenesis and provides a clinically translatable seven-gene diagnostic model for early osteoporosis detection. The integration of multiomics analyses uncovered key pathways, immune dynamics, and regulatory networks, while experimental validation reinforced the role of specific ERSRGs. These findings provide novel insights into ERS-mediated mechanisms and therapeutic targets for osteoporosis management.