Maternal Vitamin D-Binding Protein, Vitamin D levels at birth, VDR Genotype and Risk of Childhood Type 1 Diabetes

Background: Vitamin D circulates mainly as 25-hydroxyvitamin D (25OHD) bound to vitamin D-binding protein (DBP), and may influence immunity. DBP concentration doubles during pregnancy, and higher maternal DBP concentration during late pregnancy was associated with lower risk of childhood type 1 diabetes (T1D) in a previous independent study. DBP may influence plasma half-life and/or bioavailability of 25OHD, and may itself infleunce immunity (e.g macrophage activation). Furthermore, maternal 25OHD has shown inconsistent association with childhood T1D, and the association may depend on genetic variants in the vitamin D receptor (VDR).

Methods: This study included 189 mother/child pairs where the child later developed T1D, and 576 mother/child dyads as controls, randomly selected within the Norwegian Mother and Child Cohort Study. Plasma 25OHD and DBP were measured using LC-MS/MS and radioimmunoassay, respectively, in maternal samples collected mid-pregnancy and 1-3 days postpartum, and in cord blood at birth. We genotyped the mother and child for a single nucleotide polymorphism (SNP) in each of seven genes involved in vitamin D metabolism (GC, DHCR7, CYP2R1, CYP24A1, CYP27B1, VDR). We adjusted for season of samplung, child`s HLA genotype, maternal ethnicity, age and T1D in logistic regression.

Results: Higher maternal DBP levels at delivery were associated with lower offspring T1D risk (OR=0.83, 95%CI 0.70-0.98, per µmol/L increase). There was an interaction between 25OHD in cord blood and child`s VDR genotype, with higher 25OHD levels significantly associated with lower T1D risk (OR=0.84, 95%CI 0.72-0.997 per 10nmol/L increase) in children carrying the VDR SNP rs11568820 GG genotype (P-interaction=0.01). Mutually adjusting 235OHD and DBP did not appreciable change estimates. No other significant interactions were found between SNPs and 25OHD or DBP. Our results suggest that perinatal vitamin D metabolism may influence childhood T1D risk, although the question of causality and mechanisms involved remains unclear.