Identification of novel CSF biomarkers for neurodegeneration and their validation by a high-throughput multiplexed targeted proteomic assay

Background: Currently there are no effective treatments for many neurodegenerative diseases. Reliable biomarkers for identifying and stratifying these diseases will be important in the development of future novel therapies. Lewy Body Dementia (LBD) is considered an under diagnosed form of dementia for which markers are needed to discriminate LBD from other forms of dementia such as Alzheimer's Disease (AD). This work describes a Label-Free proteomic profiling analysis of cerebral spinal fluid (CSF) from non-neurodegenerative controls and patients with LBD. Using this technology we identified several potential novel markers for LBD. These were then combined with other biomarkers from previously published studies, to create a 10 min multiplexed targeted and translational MRM-LC-MS/MS assay. This test was used to validate our new assay in a larger cohort of samples including controls and the other neurodegenerative conditions of Alzheimer's and Parkinson's disease (PD). Results: Thirty eight proteins showed significantly (p < 0.05) altered expression in LBD CSF by proteomic profiling. The targeted MRM-LC-MS/MS assay revealed 4 proteins that were specific for the identification of AD from LBD: ectonucleotide pyrophosphatase/phosphodiesterase 2 (p < 0.0001), lysosome-associated membrane protein 1 (p < 0.0001), pro-orexin (p < 0.0017) and transthyretin (p < 0.0001). Nineteen proteins were elevated significantly in both AD and LBD versus the control group of which 4 proteins are novel (malate dehydrogenase 1, serum amyloid A4, GM2-activator protein, and prosaposin). Protein-DJ1 was only elevated significantly in the PD group and not in either LBD or AD samples. Correlations with Alzheimer-associated amyloid beta-42 levels, determined by ELISA, were observed for transthyretin, GM2 activator protein and IGF2 in the AD disease group (r(2) >= 0.39, p <= 0.012). Cystatin C, ubiquitin and osteopontin showed a strong significant linear relationship (r(2) >= 0.4, p <= 0.03) with phosphorylated-tau levels in all groups, whilst malate dehydrogenase and apolipoprotein E demonstrated a linear relationship with phosphorylated-tau and total-tau levels in only AD and LBD disease groups. Conclusions: Using proteomics we have identified several potential and novel markers of neurodegeneration and subsequently validated them using a rapid, multiplexed mass spectral test. This targeted proteomic platform can measure common markers of neurodegeneration that correlate with existing diagnostic makers as well as some that have potential to show changes between AD from LBD.