Biochemical Profiling of the Brain and Blood Metabolome in a Mouse Model of Prodromal Parkinson's Disease Reveals Distinct Metabolic Profiles

Stewart F. Graham, Beaumont Health
Nolwen L. Rey, Van Andel Research Institute
Ali Yilmaz, Beaumont Health
Praveen Kumar, Beaumont Health
Zachary Madaj, Van Andel Research Institute
Michael Maddens, Beaumont Health
Ray O. Bahado-Singh, Beaumont Health
Katelyn Becker, Van Andel Research Institute
Emily Schulz, Van Andel Research Institute
Lindsay K. Meyerdirk, Van Andel Research Institute
Jennifer A. Steiner, Van Andel Research Institute
Jiyan Ma, Van Andel Research Institute
Patrik Brundin, Van Andel Research Institute

Abstract

© 2018 American Chemical Society. Parkinson's disease is the second most common neurodegenerative disease. In the vast majority of cases the origin is not genetic and the cause is not well understood, although progressive accumulation of α-synuclein aggregates appears central to the pathogenesis. Currently, treatments that slow disease progression are lacking, and there are no robust biomarkers that can facilitate the development of such treatments or act as aids in early diagnosis. Therefore, we have defined metabolomic changes in the brain and serum in an animal model of prodromal Parkinson's disease. We biochemically profiled the brain tissue and serum in a mouse model with progressive synucleinopathy propagation in the brain triggered by unilateral injection of preformed α-synuclein fibrils in the olfactory bulb. In total, we accurately identified and quantified 71 metabolites in the brain and 182 in serum using 1 H NMR and targeted mass spectrometry, respectively. Using multivariate analysis, we accurately identified which metabolites explain the most variation between cases and controls. Using pathway enrichment analysis, we highlight significantly perturbed biochemical pathways in the brain and correlate these with the progression of the disease. Furthermore, we identified the top six discriminatory metabolites and were able to develop a model capable of identifying animals with the pathology from healthy controls with high accuracy (AUC (95% CI) = 0.861 (0.755-0.968)). Our study highlights the utility of metabolomics in identifying elements of Parkinson's disease pathogenesis and for the development of early diagnostic biomarkers of the disease.