Biomarkers of total selenium and selenium species in paired serum and cerebrospinal fluid samples

Background: The validity of biomarkers to estimate exposure to selenium (Se) species and selenoproteins in the central nervous system (CNS) is not well studied. Methods: Among 83 Italian participants with mild cognitive impairment, we estimated total Se and single Se species concentrations in paired serum and cerebrospinal fluid (CSF) samples using anion exchange chromatography-inductively coupled plasma-dynamic reaction cell-mass spectrometry. In each matrix (serum and CSF), we assessed associations between: 1) paired Se species and 2) total Se and Se species. Results: The distribution of Se exposure was comparable to that generally found in European populations. We found few consistent patterns for most biomarkers, including total Se and some Se species. An exception was a positive association between the two matrices for selenoprotein-P-bound Se and the inorganic Se form selenate, and an unexpected inverse association for glutathione-peroxidase-bound Se. Total Se was positively associated with some Se species but inversely associated with other Se species in serum, while in CSF the positive association was stronger and more consistent across various Se species. Conclusions: Concentrations of total Se and single Se species in serum were not strongly correlated with their respective concentrations in CSF, the gold standard to estimate CNS exposure. Furthermore, total Se and selected Se species showed consistent positive correlations within CSF but not serum. Our results suggest that relying on serum Se concentrations to assess CNS exposure can introduce error in human studies.