Targeted Metabolomic Analysis of Soluble Lysates from Platelets of Patients with Mild Cognitive Impairment and Alzheimer’s Disease Compared to Healthy Controls: Is PC aeC40:4 a Promising Diagnostic Tool?
Affiliations: [a] Department of Legal Medicine and Core Facility Metabolomics, Medical University of Innsbruck, Austria
| [b] Department of Medical Statistics, Informatics and Health Economics, Medical University of Innsbruck, Austria
| [c] Department of Psychiatry and Psychotherapy A, Hall State Hospital, Austria
| [d] Laboratory of Psychiatry and Experimental Alzheimer’s Research, Medical University of Innsbruck, Austria
Correspondence:
[*]
Correspondence to: Christian Humpel, PhD, Department of Psychiatry and Psychotherapy, Anichstr. 35, A-6020 Innsbruck, Austria. Tel.: +43 512 504 23712; Fax: +43 512 504 23713; E-mail: [email protected].
Abstract: Alzheimer’s disease (AD) is a progressive neurodegenerative disorder of the central nervous system. The use of biological fluids in AD diagnosis remains limited to the analysis of specific protein biomarkers in cerebrospinal fluid. However, metabolomic analysis has recently revealed several metabolites in plasma, especially phosphatidylcholines (PC), as putative biomarkers specific for AD. Following on previous reports of platelet abnormalities in AD, we hypothesized that platelets metabolites released in plasma may offer new biomarkers in AD. The aim of the present study was to apply targeted metabolomics to compare metabolites in soluble lysates of platelets from healthy controls (CO), patients with mild cognitive impairment (MCI), and patients with AD in a cohort of 90 subjects. We could target 163 metabolites and quantitative data were obtained for 91 metabolites. Among these, the lipid PC aeC40:4 significantly differentiated AD from CO (p = 0.0009), while four other lipids (PC aaC32:0, PC ae C32:2, PC aeC34:1, and SM(OH)C14:1) differentiated patients with MCI from CO. The combination of three phosphatidylcholines (PC aeC32:2, PC aeC34:1, PCaaC36:5), two lyso-phosphatidylcholines (lysoPC aC18:1, lysoPC aC16:0), and one sphingomyelin (SM(OH) C14:1) constructed a valuable prediction model using the C4.5 decision tree. The diagnosis accuracy for AD versus CO and MCI was 85%. In a blinded follow up conversion study (10 patients with a second blood collection after 9 months), we could verify the clinical diagnosis in 19 out of 20 cases. We propose that soluble platelet PCaeC40:4 is a promising marker to diagnose AD with a cut-off of <0.30μM and that platelets undergo metabolic processes during AD progression.
