Predictors of selenium biomarker kinetics in 4–9-year-old Bangladeshi children

Abstract

Background: Biomarker selenium concentrations vary greatly between studies. Concentrations in erythrocytes, urine, and hair vary even at similar plasma concentrations, suggesting that unknown factors influence the distribution of selenium between body compartments. Objective: To assess predictors of the different selenium biomarkers in children. Design: We used a mother-child cohort, nested in a population-based supplementation trial in rural Bangladesh (MINIMat), established for evaluation of arsenic toxicity. Selenium was measured in plasma (n = 223), erythrocytes, urine, and hair at 9 years (n = 395) and in erythrocytes and urine at 4.5 years (n = 259) using inductively coupled plasma mass spectrometry. We also measured concentrations of arsenic (all biospecimen) and cadmium (erythrocytes and urine). Genotyping for INMT, a methyltransferase involved in selenium metabolism, was performed using TaqMan probes. Results: At 9 years, the selenium concentrations ranged 51–139 μg/L in plasma, 128–281 μg/L in erythrocytes, 2.2–55 μg/L in urine, and 258–723 μg/kg in hair. Correlations (rS) between biomarkers ranged 0.12–0.37, and were strongest between blood compartments and between erythrocytes and hair (long-term markers). In multivariable-adjusted linear regression analyses, plasma selenium differed by sampling season (highest in food-secure pre-monsoon season) and was inversely associated with plasma arsenic (range < 0.0080–20 μg/L; B = −1.1, 95% CI: -1.8, −0.41). In contrast, erythrocyte selenium was positively associated with erythrocyte arsenic (range 0.95–50 μg/L; B = 0.58, 95% CI: 0.26, 0.91) and inversely associated with erythrocyte cadmium (range 0.27–3.1 μg/L; B = −12, 95% CI: −17, −6.9). These associations were similar at 4.5 years. Only selenium in hair and urine were influenced by INMT polymorphisms. Finally, chronic malnutrition seemed to increase selenium retention, measured as the ratio plasma/urinary selenium. Conclusions: Selenium biomarkers seem to be influenced by malnutrition, genetics, and exposure to metal pro-oxidants. This might affect the evaluation of deficiency/sufficiency, normally assessed by selenium in plasma/serum. Keywords: Metabolism, Se, Red blood cells, Biospecimen, Serum, Metals