Serum metabolomic biomarkers of perceptual speed in cognitively normal and mildly impaired subjects with fasting state stratification

Authors: BORKOWSKI, KAMIL - University Of California, Davis; TAHA, AMEER - University Of California, Davis; PEDERSEN, THERESA - University Of California, Davis; DE JAGER, PHILIP - Columbia University Medical Center; BENNETT, DAVID - Rush University; ARNOLD, MATTHIAS - Duke University; KADDURAH-DAOUK, RIMA - Duke University; Newman, John

Submitted to: Scientific Reports
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/5/2021
Publication Date: 9/23/2021
Citation: Borkowski, K., Taha, A.Y., Pedersen, T.L., De Jager, P.L., Bennett, D.A., Arnold, M., Kaddurah-Daouk, R., Newman, J.W. 2021. Serum metabolomic biomarkers of perceptual speed in cognitively normal and mildly impaired subjects with fasting state stratification. Scientific Reports. 11. Article 18964. https://doi.org/10.1038/s41598-021-98640-2.
DOI: https://doi.org/10.1038/s41598-021-98640-2

Interpretive Summary: Cognitive decline is associated with both normal aging and early pathologies leading to dementia. Metabolic biomarkers of cognitive decline may provide insight into mechanisms of neurocognitive disorder development. Neurocognitive impairments are linked to inflammation and changes in vascular regulation. In this study we explored the serum profiles of a suite of small molecules important in the regulation of inflammation, neural function, energy metabolism and vascular function for biomarkers of cognitive decline. Serum samples (n =210) were obtained from the Religious Order Study/Memory and Aging Project (ROSMAP) from subjects with either mild cognitive impairment (MCI) or no cognitive impairment (NCI). Samples were collected opportunistically, and information regarding the fasting state which can compromise biomarker discovery was incomplete. Hence, subjects were stratified into fasted (n =59), non-fasted (n = 80), and unknown fasting states (n = 71), and we developed a predictive model to estimate the state of unknown subjects and stratify our analysis. Using this approach, we found distinct biomarkers in the fasted and non-fasted states. In the non-fasted state, we identified positive associations between perceptual speed and serum levels of circulating linoleic acid and palmitoleoyl ethanolamide, a lipid associated with adipose tissue function. In the fasted state, we identified negative associations between perceptual speed and soluble epoxide hydrolase activity, and important regulator of vascular constriction, and a low abundance bile acid metabolite of unknown function. In the non-fasted state, cognitive domains other than perceptual speed showed negative associations with conjugated to unconjugated bile acids ratio and levels of bile acids derived from gut microbial metabolism. Importantly, this study provides a predictive model for the fasting state based on measureable metabolites in serum samples and shows unique associations between serum metabolites and perceptual speed in the fasted and non-fasted samples. These results demonstrate the utility of serum metabolomics for cognitive biomarker discovery.

Technical Abstract: Cognitive decline is associated with both normal aging and early pathologies leading to dementia. Here we used quantitative profiling of lipid mediators including oxylipins, endocannabinoids, bile acids, and steroid hormones to investigate associations with cognitive domains. Serum samples (n =210) were opportunistically collected from subjects with or without MCI. To maximize power and stratify the analysis by the fasting state, we developed an algorithm to predict subject fasting state when unknown (n =71). In non-fasted subjects, linoleic acid and palmitoleoyl ethanolamide levels were positively associated with perceptual speed. In fasted subjects, soluble epoxide hydrolase activity and tauro-alpha-muricholic acid levels were negatively associated with perceptual speed. Other cognitive domains showed associations with bile acid metabolism only in the non-fasted state. Importantly, this study shows unique associations between serum metabolites and cognitive function in the fasted and non-fasted states and provides a fasting state prediction algorithm based on quantifiable metabolites.