Metabolomic biomarkers of rest-activity rhythms in older men: results from the Osteoporotic Fractures in Men (MrOS) study.

STUDY OBJECTIVES: Rest-activity rhythm characteristics have been linked to a wide range of health conditions; however, the molecular mechanisms underlying these associations are not well understood. This study is the first of two studies aiming to use an untargeted approach to identify metabolomic markers associated with rest-activity rhythm characteristics and focuses on older men. METHODS: The study included 950 participants from the Osteoporotic Fractures in Men study. Multiple parametric and non-parametric variables of rest-activity rhythms were derived from actigraphy data. A total of 848 metabolites were measured from fasting blood samples using an untargeted approach. Multiple linear regression models and Ingenuity Pathway Analysis (IPA) were used to identify metabolomic profiles associated with rest-activity variables. RESULTS: We found 65 metabolites, mostly amino acids and lipids, that were significantly associated with at least one of the primary rest-activity variables (i.e. pseudo-F-statistic, intradaily variability, and interdaily stability). These metabolites were from various biochemical pathways, including diacylglycerol, plasmalogen, lysoplasmalogen, and amino sugar metabolism. The IPA suggested that these metabolites may be implicated in various diseases and functions, particularly immune and inflammatory diseases, and identified the PEX2-PEX5 network as a significantly enriched gene-regulation pathway. CONCLUSIONS: Our findings expand the current knowledge about the relationship between diurnal behaviors and human metabolism, and provide new evidence regarding mechanistic pathways that may mediate the adverse health effects of impaired rest-activity rhythms in older men. Statement of Significance In this metabolomics study in older men, we found a large number of metabolites that were associated with rest-activity rhythms. Our findings expand the current knowledge about the relationship between circadian-regulated diurnal behaviors and human metabolism, reinforce the critical role of circadian function in health and diseases, and provide new evidence regarding mechanistic pathways that may mediate the adverse effects of circadian disruptions. Our findings also point to ample future directions for further research to further elucidate the relationships among rest-activity rhythms, metabolomic profiles, and disease risk, which may help identifying intermediate targets for developing disease therapies and developing models for disease risk prediction and management.