GPX1 and RCN1 as New Endoplasmic Reticulum Stress-Related Biomarkers in Multiple Sclerosis Brain Tissue and Their Involvement in the APP-CD74 Pathway: An Integrated Study Combining Machine Learning and Multi-Omics.

This study identified 13 endoplasmic reticulum stress (ERS)-related biomarkers associated with multiple sclerosis (MS) through integrated bioinformatics analysis (including weighted gene co-expression network analysis and machine learning algorithms) and single-cell sequencing, combined with validation in an experimental autoimmune encephalomyelitis (EAE) mouse model. Among them, GPX1, RCN1, and UBE2D3 exhibited high diagnostic value (AUC > 0.7, p < 0.05), and the diagnostic potential of GPX1 and RCN1 was confirmed in the animal model. The study found that memory B cells, plasma cells, neutrophils, and M1 macrophages were significantly increased in MS patients, while naive B cells and activated NK cells decreased. Consensus clustering based on key ERS-related genes divided MS patients into two subtypes. Single-cell sequencing showed that microglia and pericytes were the cell types with the highest expression of key ERS-related genes, and the APP-CD74 pathway was enhanced in the brain tissue of MS patients. Mendelian randomization analysis suggested that GPX1 plays a protective role in MS. These findings reveal the mechanisms of ERS-related biomarkers in MS and provide potential targets for diagnosis and treatment.