Plasma Levels of Food-Derived Metabolites as Biomarkers of Parkinson's Disease.
Xiaoxue Dong, Yilong Zheng, Evelyn Ting Ying Tan, Qiao Yang Sun, Bin Xiao, Eng King Tan, Yun-Cheng Wu, Zhi Dong Zhou
Abstract
Open AccessParkinson's disease (PD) is a progressive neurodegenerative disorder shaped by genetic factors such as LRRK2 and GBA1 mutations, as well as dietary and metabolic influences. Food-derived plasma metabolites-including caffeine, paraxanthine, trigonelline, piperine, and sitosteryl hexoside-have emerged as promising, accessible biomarkers for early detection, progression monitoring, and therapeutic targeting, yet their longitudinal behavior and genetic interactions remain insufficiently characterized. Using the Parkinson's Progression Markers Initiative (PPMI) cohort (n = 455; 303 PD patients, 152 controls), we quantified plasma levels of these metabolites by quantitative LC-MS/MS with batch correction, examining sporadic PD and genetically defined subgroups (LRRK2-PD [PDL], GBA1-PD [PDG], dual-mutation PD [PDGL], and prodromal equivalents). Baseline one-way ANOVA showed significantly lower caffeine and paraxanthine in PDL (p = 0.0467, p = 0.0178) and PDG (p = 0.0408), reduced piperine in PDL (p = 0.0009), PDG (p = 0.0257), and prodromal LRRK2 (p = 0.0168), and elevated sitosteryl hexoside in PDG (p = 0.0184). Longitudinal regression analyses revealed that in sporadic PD, caffeine negatively correlated with MDS-UPDRS parts I (β = -2, p = 0.0475) and III (β = -7.2, p = 0.007), trigonelline declined over time and was inversely associated with part III (β = -1.7, p = 0.0069), and sitosteryl hexoside negatively correlated with parts II (β = -68.3, p = 0.042) and III (β = -74.1, p = 0.0425). In PDL, sitosteryl hexoside inversely correlated with part I (β = -54.2, p = 0.0049), while in PDGL, paraxanthine showed negative associations with part II (β = -18.5, p = 0.00327). These findings demonstrate subgroup-specific alterations in food-derived metabolites and consistent inverse associations with PD severity, supporting their potential as non-invasive biomarkers, particularly in LRRK2/GBA1 mutation carriers, and highlighting the need for longitudinal validation and dietary intervention trials to advance personalized PD management.