PiB-PET Imaging-Based Serum Proteome Profiles Predict Mild Cognitive Impairment and Alzheimer's Disease

Development of a simple, non-invasive early diagnosis platform of Alzheimer's disease (AD) using blood is urgently required. Recently, PiB-PET imaging has been shown to be powerful to quantify amyloid-β plaque loads leading to pathophysiological alterations in AD brains. Thus, there has been a need for serum biomarkers reflecting PiB-PET imaging data as an early diagnosis platform of AD. Here, using LC-MS/MS analysis coupled with isobaric tagging, we performed comprehensive proteome profiling of serum samples from cognitively normal controls, mild cognitive impairment (MCI), and AD patients, who were selected using PiB-PET imaging. Comparative analysis of the proteomes revealed 79 and 72 differentially expressed proteins in MCI and AD, respectively, compared to controls. Integrated analysis of these proteins with genomic and proteomic data of AD brain tissues, together with network analysis, identified three biomarker candidates representing the altered proteolysis-related process in MCI or AD: proprotein convertase subtilisin/kexin type 9 (PCSK9), coagulation factor XIII, A1 polypeptide (F13A1), and dermcidin (DCD). In independent serum samples of MCI and AD, we confirmed the elevation of the candidates using western blotting and ELISA. Our results suggest that these biomarker candidates can serve as a potential non-invasive early diagnosis platform reflecting PiB-PET imaging for MCI and AD.