Relationships between metabolic activity in the brain and circulating metabolites vary by sex

Abstract

AbstractBackgroundGlucose hypometabolism in the brain measured by [F‐18] fluorodeoxyglucose (FDG) positron emission tomography has been observed in a number of neurological disorders. In both Down syndrome (DS) and non‐DS populations, it is associated with higher amyloid load and lower cognitive performance.MethodWe obtained plasma samples, FDG PET and [C‐11] Pittsburgh compound B (PiB) from a sample of 89 adults with DS (mean age [standard deviation] = 38.9[8.3] years). We measured levels of metabolites using targeted mass spectrometry. We calculated correlation constants between glucose uptake in discrete brain regions, global amyloid load, and plasma metabolite levels. We examined the effect of sex and age on these correlations.ResultWe found significant correlations between plasma metabolites, primarily lipids, and FDG PET signal in the brain with regional specificity. Additionally, metabolites which were positively associated with FDG PET signal in a specific brain region were negatively associated with amyloid load and vice versa. While FDG PET signal in each region were not significantly different between males and females, their correlation with metabolites did. When examining females only, we find primarily lipid correlates again. The lipid metabolites negatively associated with FDG PET signal were distinguished from those lipids positively correlated with FDG PET signal by lipid class and not fatty acid length. In males, a completely distinct set of associations between brain regions and metabolites were observed. The number of FDR significant correlations decrease dramatically, however the magnitudes of these correlation constants are higher than those observed in females and there is very little overlap between any of the brain regions correlated with each metabolite.ConclusionThe levels of brain amyloid and regional brain glucose metabolism in adults with DS are both reflected in the concentration of certain metabolites measured in blood plasma. Surprisingly, each brain region has its own orthogonal metabolic correlate. Moreover, these relationships are distinct in males and females. It will be important to explore the basis for these differences and determine if they are present in the general population. Future metabolic analysis of neurological diseases could also benefit from separate analysis in males and females so as not to obscure these differences.