Abstract
To our knowledge, this study may be the first to examine the antagonistic role of selenium (Se) on oxidative stress induced by cadmium (Cd) and its impact on birth measures. Cd and Se levels were measured in umbilical-cord blood and the placentas of a subsample of 250 healthy mothers who participated between 2005 and 2006 in the project "Prenatal Exposure to Pollutants". The median Cd levels in cord and maternal blood and placental tissue were 0.78μg/l, 0.976μg/l and 0.037μg/g dry wt., respectively. The median levels of Se in cord serum and placental tissue were 65.68μg/l and 1.052μg/g dry wt., respectively. Se was more than 100-fold in molar excess over Cd in both cord serum and placental tissue. The median molar Cd/Se ratios in cord serum and placental tissue were 0.008 and 0.024, respectively, which were much lower than unity. This study suggests that both Cd and Se play a role in the mechanism of oxidative stress, but, the process underlying this mechanism remains unclear. Nevertheless, three biomarkers of oxidative stress had inconsistent relationships with Cd and/or Se in various matrices, perhaps due to potential untested confounders. Our results generally support an association between low in utero exposure to Cd and the anthropometric development of the fetus. Adjusted regression models indicated a negative association of cord blood Cd levels ≥0.78μg/l with Apgar 5-min scores and birth height. Maternal Cd levels ≥0.976μg/l were associated with a 5.94-fold increased risk of small-for-gestational-age births, which increased to 7.48-fold after excluding preterm births. Placenta weight decreased with increasing placental Cd levels ≥0.037μg/g dry wt. (p=0.045), an association that became stronger after excluding preterm births or adjusting for birth weight. Cord Se levels ≥65.68μg/l were positively associated with placenta weight (p=0.041) and thickness (p=0.031), an association that remained unchanged after excluding preterm births. Cord Se levels, however, were negatively associated with cephalization index, but only after excluding preterm births (p=0.017). Each birth measure was again modeled as a function of the Cd/Se ratios in cord blood and placenta tissue. Interestingly cord ratios ≥0.008 were negatively associated with Apgar-5min score (p=0.047), birth weight (p=0.034) and placenta thickness (p=0.022). After excluding preterm births, only the association with placenta thickness remained significant (p=0.021), while birth weight (p=0.053) was marginally significant. In contrast, cephalization index increased with Cd/Se ratios ≥0.008 (p=0.033), an association that became marginally significant after excluding preterm births (p=0.058). For placental Cd/Se ratios ≥0.024, only placenta weight was reduced with (p=0.037) and without (p=0.009) the inclusion of preterm births. These findings do not support an antagonistic mechanism between Cd and Se. The role of oxidative mechanisms either induced by Cd exposure or alleviated by Se on these birth anthropometric measures was examined by principal component analysis. Se did not have a clear protective role against Cd-induced adverse effects despite its substantial excess over Cd, and its role in alleviating oxidative stress by reducing malondialdehyde levels. The results may suggest that the extent of the Se beneficial effects is not governed only by its concentration but also by the chemical forms of Se that interact with various proteins. Consequently, the speciation of Se in such studies is essential for understanding and predicting Se availability for absorption.