Abstract
BACKGROUND: Given the association between iron deficiency and lead absorption, we hypothesized that variants in iron metabolism genes would predict higher blood lead levels in young children. OBJECTIVE: We examined the association between common missense variants in the hemochromatosis (HFE) and transferrin (TF) genes and blood lead levels in 422 Mexican children. METHODS: Archived umbilical cord blood samples were genotyped for HFE (H63D and C282Y) and TF (P570S) variants. Blood lead was measured at 24, 30, 36, 42, and 48 months of age. A total of 341 subjects had at least one follow-up blood lead level available and data available on covariates of interest for inclusion in the longitudinal analyses. We used random-effects models to examine the associations between genotype (HFE, TF, and combined HFE + TF) and repeated measures of blood lead, adjusting for maternal blood lead at delivery and child's concurrent anemia status. RESULTS: Of 422 children genotyped, 17.7, 3.3, and 18.9% carried the HFE H63D, HFE C282Y, and TF P570S variants, respectively. One percent of children carried both the HFE C282Y and TF P570S variants, and 3% of children carried both the HFE H63D and TF P570S variants. On average, carriers of either the HFE (beta = 0.11, p = 0.04) or TF (beta = 0.10, p = 0.08) variant had blood lead levels that were 11% and 10% higher, respectively, than wild-type subjects. In models examining the dose effect, subjects carrying both variants (beta = 0.41, p = 0.006) had blood lead 50% higher than wild-type subjects and a significantly higher odds of having a blood lead level > 10 microg/dL (odds ratio = 18.3; 95% confidence interval, 1.9-177.1). CONCLUSIONS: Iron metabolism gene variants modify lead metabolism such that HFE variants are associated with increased blood lead levels in young children. The joint presence of variant alleles in the HFE and TF genes showed the greatest effect, suggesting a gene-by-gene-by-environment interaction.