Transcriptional Landscape and Biomarker Discovery for Endoplasmic Reticulum Stress in Alzheimer's Disease: An Ex Vivo Study Using Patients-Derived Dermal Fibroblasts.

OBJECTIVE: Numerous studies have identified various risk factors associated with Alzheimer's disease (AD). However, the experimental limitations of disease modeling make it challenging to directly interpret their effects. These limitations include constraints of postmortem samples, animal experiments, and challenges associated with brain tissue studies. Ex vivo experiments effectively address these issues by enabling patient-specific identification and highlighting potential biomarkers. This study aimed to characterize the transcriptional profile of fibroblasts derived from patients with AD in response to endoplasmic reticulum (ER) stress and propose potential biomarkers. METHODS: We utilized an ex vivo platform to identify genes differentially responsive to ER stress. The transcriptional feature of fibroblasts in both healthy controls (n=22) and patients with AD (n=20) was analyzed using bulk RNA sequencing. The cytotoxicity of the selected target gene was evaluated through knockdown experiments. RESULTS: A total of 468 differentially expressed genes (DEGs) were identified. Gene ontology and pathway enrichment analysis revealed that 210 DEGs, which were less responsive in AD, are involved in lipid-related terms and pathways. By narrowing down AD-related genes, we identified 49 highly reliable AD-associated genes. The most significant gene, DCTN2, exhibited a fold change that positively correlated with cognitive function and negatively correlated with blood-based biomarkers (pTau217, amyloid beta 42/40 ratio), aligning with the amyloid/Tau/neurodegeneration research criteria for AD. Additionally, the knockdown of DCTN2 in glial cell lines resulted in increased cell toxicity and apoptosis. CONCLUSION: Identifying differentially responsive genes in ex vivo experiments not only provides insights into the pathology of AD but also offers potential biomarkers for disease diagnosis.