Premium
Serum metabolome informs neuroimaging biomarkers for Alzheimer’s disease
Author(s) -
Nho Kwangsik,
KueiderPaisley Alexandra,
Arnold Matthias,
Dehkordi Siamak Mahmoudian,
Risacher Shan L.,
Louie Gregory,
Blach Colette,
Baillie Rebecca,
Han Xianlin,
Kastenmüller Gabi,
Doraiswamy P. Murali,
KaddurahDaouk Rima F.,
Saykin Andrew J.
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.045596
Subject(s) - neurodegeneration , neuroimaging , metabolome , alzheimer's disease neuroimaging initiative , atrophy , alzheimer's disease , metabolite , medicine , amyloid (mycology) , biomarker , disease , cognitive decline , neuroscience , endocrinology , pathology , biology , dementia , biochemistry
Background Accumulating evidence suggests that Alzheimer’s disease (AD) patients display metabolic dysfunction. Metabolomics is a powerful tool for studying metabolic pathways by mapping biochemical changes in AD, yet little is known about the association between circulating metabolites in blood and the “A/T/N” (amyloid‐β, tau, neurodegeneration) biomarkers for AD. Method Serum‐based targeted metabolite levels were measured in the Alzheimer’s Disease Neuroimaging Initiative (ADNI; n = 1,531; 370 cognitively normal controls (CN), 95 significant memory concern (SMC), 271 early mild cognitive impairment (EMCI), 491 late MCI (LMCI), and 304 AD). The quality control process resulted in 139 metabolites for further analysis. We performed association analyses of circulating serum metabolites with the “A/T/N” biomarkers for AD. Biomarkers of “A” are CSF Aβ1‐42 and cortical amyloid‐β accumulation measured by Florbetapir PET, biomarkers of “T” are CSF phosphorylated tau (p‐tau), and biomarkers of “N” are atrophy on MRI and glucose metabolism on FDG PET. FDR‐based multiple comparison adjustment with the Benjamini‐Hochberg procedure was used. Result We found significant associations of multiple metabolites with biomarkers of amyloid‐β (“A”) and neurodegeneration (“N”) including brain amyloid deposition, glucose metabolism, and structural atrophy measured from multimodal neuroimaging scans (MRI, PET) after multiple testing adjustment. We did not observe any significant associations between metabolites and biomarkers of fibrillary tau (“T”) (Figure 1). In particular, higher levels of phosphatidylcholines (PC) were associated with greater amyloid deposition (“A”) and decreased brain glucose metabolism (“N”). Higher levels of propionyl‐L‐carnitine (C3) were associated with less amyloid deposition and increased brain glucose metabolism. Higher levels of two amino acids, two PCs, and three sphingomyelins were associated with less brain atrophy (“N”). Conclusion This is the first study to show serum circulating metabolites are significantly associated with neuroimaging biomarkers for AD. Perturbations in PC and acylcarnitine metabolism may play a role in features intrinsic to AD including amyloid‐β deposition and neurodegeneration, but not fibrillary tau suggesting that acylcarnitines and PCs are associated with earlier stages of AD as tau spreading is more closely related to AD progression.