Common Variants and Haplotypes in the TF, TNF-α, and TMPRSS6 Genes Are Associated with Iron Status in a Female Black South African Population1–3

Abstract

Background: It is unknown whether single nucleotide polymorphisms (SNPs), associated with iron status in European and Asian populations, have the same relation within the African population.Objectives: We investigated associations of reported SNPs with iron markers in a South African cohort.Methods: Hemoglobin concentration, serum ferritin (SF) and soluble transferrin receptor (sTfR) concentrations, and body iron (BI) stores were measured in women (n = 686; range, 32–86 y) who were part of the Prospective Urban and Rural Epidemiology study. Thirty-two SNPs in 12 genes were selected based on existing genome-wide association study data.Results: In the transferrin (TF) gene, SF and BI were significantly lower in the heterozygote genotype (AG) of reference SNP (rs) 1799852 (P = 0.01 and 0.03, respectively) and sTfR concentrations were significantly higher (P = 0.004) than the homozygote minor allele genotype (AA), whereas transferrin receptor and BI concentrations were significantly lower in the heterozygote genotype (AG) of rs3811647 (both P = 0.03) than the homozygote wild-type (AA) and minor allele groups (GG). The chromosome 6 allele combination (AAA) consisting of rs1799964 and rs1800629 both in tumor necrosis factor-α (TNF-α) and rs2071592 in nuclear factor κB inhibitor–like protein 1 (NFKBIL1) was associated with higher odds for low SF concentrations (SF < 15 μg/L; OR: 1.86; 95% CI: 1.23, 2.79) than the allele combinations AGA, GGT, and AGT. The chromosome 22 allele combination (GG) consisting of rs228918 and rs228921 in the transmembrane protease serine 6 (TMPRSS6) gene was associated with lower odds for increased sTfR concentrations (sTfR > 8.3mg/L; OR: 0.79; 95% CI: 0.63, 0.98) than the allele combination AA.Conclusions: Various SNPs and allele combinations in the TF, TNF-α, and TMPRSS6 genes are associated with iron status in black South African women; however, these association patterns are different compared with European ancestry populations. This stresses the need for population-specific genomic data.