作者
Rebecca G. Smith,Isabelle Bos,Stephanie J.B. Vos,Frans R.J. Verhey,Philip Scheltens,Sebastiaan Engelborghs,Giovanni B. Frisoni,Olivier Blin,Jill C. Richardson,Régis Bordet,Magda Tsolaki,Julius Popp,Pablo Martínez-Lage,Alberto Lleó,Peter Johannsen,Yvonne R. Freund,Lutz Frölich,Rik Vandenberghe,Simon Lovestone,Johannes Streffer,Ulf Andréasson,Kaj Blennow,Pieter Jelle Visser,Henrik Zetterberg,Lars Bertram,Katie Lunnon
摘要
Abstract Background Alzheimer’s disease is associated with increases in amyloid β and hyperphosphorylated tau which is thought to occur decades before the onset of clinical symptoms, leading to cell loss and inhibition. Finding biomarkers to detect these changes before neuronal loss and therefore permanent damage has occurred is integral. Current biomarkers using cerebrospinal fluid (CSF) measures and positron‐emission tomography (PET) imaging are invasive or expensive to perform. In this study we use the European Medical Information Framework Alzheimer’s disease (EMIF‐AD) cohort to looks for blood related epigenetic (DNA methylation) changes which relate to CSF and existing peripheral biomarkers. Method Whole blood samples from EMIF‐AD were accessed for DNA methylation on Illumina EPIC arrays. We assessed DNA methylation levels in 886 individuals with > 10 central and local biomarker measures, including amyloid, tau, neurogranin and YKL‐40. We also performed regional analysis to look for regions of significant DNA methylation change and identified specific pathways that show methylomic changes in association with specific biomarkers. Result We identified epigenome‐wide significant probes associated with many of our biomarker measures including most notably central amyloid β‐42, amyloid β‐40/42 ratio, neurogranin and YKL‐40 CSF levels as well as with local phosphorylated tau and total tau measurements. We also are able to use collections of CpG sites to create classifiers that predict high levels of CSF biomarkers. Conclusion In the future, identified sites and collections of sites could be used as a proxy for other biomarker measures due to the less invasive and less expensive nature of blood collection DNA methylation analysis. We are currently undertaking further analysis of this cohort using imaging and clinical measures.