Identification and clinical validation of autophagy-related genes as potential biomarkers of Parkinson's disease and their correlation with immune infiltration.
Using public transcriptomic data and machine learning, the authors derived a four-gene autophagy-related blood signature (PRKD1, CAMP, MCOLN3, ATG9B) that discriminates Parkinson's patients from controls (model AUC 0.845) and correlates with altered immune-cell infiltration, with RT-qPCR and plasma…
What the AI sees
Using public transcriptomic data and machine learning, the authors derived a four-gene autophagy-related blood signature (PRKD1, CAMP, MCOLN3, ATG9B) that discriminates Parkinson's patients from controls (model AUC 0.845) and correlates with altered immune-cell infiltration, with RT-qPCR and plasma…
Research significance
Provides a clinically accessible autophagy–immune linked biomarker panel with modest diagnostic performance and nominates mechanistically relevant genes (autophagy/lysosomal and immune-related) for follow-up as potential therapeutic targets or stratification tools, though prospective validation and…
Source abstract
BACKGROUND: Parkinson's disease (PD) is the second most common progressive neurodegenerative disease that severely affects the quality of life and there is an urgent need to explore unique and effective diagnostic markers. The present study aimed to develop and validate a multigene combination model for the diagnosis of PD based on autophagy-related genes (ARGs) and to discover their correlation with immune infiltrating cells. METHODS: We downloaded the dataset GSE49126 from the database and retrieved ARGs from three databases. Differentially expressed ARGs were filtered by three machine algorithms and applied to construct diagnostic models. We then analyzed the differences in immune microenvironment between PD and controls utilizing ImmuCellAI and CIBERSORT, followed by investigating the correlation between markers and immune cells to better understand the molecular interactions mechanisms. Finally, the expression of key ARGs was validated in Peripheral blood mononuclear cells (PBMC) and plasma using RT-qPCR and ELISA respectively. RESULTS: A total of 28 differentially expressed ARGs were identified, and a four-gene diagnostic model (PRKD1, CAMP, MCOLN3, and ATG9B) was established, demonstrating favorable diagnostic performance (AUC = 0.845, 95% CI: 0.736-0.954). Immune infiltration analysis revealed alterations in the immune microenvironment in PD, and correlation analysis indicated significant associations between model ARGs and immune cell subsets. RT-qPCR validation showed increased expression of PRKD1, CAMP, and ATG9B in PBMCs from patients with PD compared with age- and sex-matched healthy controls (p < 0.05), consistent with transcriptomic findings. In contrast, MCOLN3 expression was decreased in patients with PD. Plasma CAMP protein levels were reduced in PD and demonstrated diagnostic potential (AUC = 0.771, 95% CI: 0.696-0.845). CONCLUSIONS: An ARG-based diagnostic model comprising PRKD1, CAMP, MCOLN3, and ATG9B demonstrated potential diagnostic value for PD and revealed associations between autophagy-related signatures and immune alterations. These findings provide exploratory evidence supporting the involvement of autophagy-immune interactions in PD and warrant further validation in larger prospective cohorts.