Objective: To determine the extent to which genetic and epigenetic factors contribute to variations in glycosylation of immunoglobulin G (IgG) in humans.
Methods: 76 N-glycan traits in circulating IgG were analyzed by UPLC in 220 monozygotic and 310 dizygotic twin pairs from TwinsUK. A classical twin study design was used to derive the additive genetic, common and unique environmental components defining the variance in these traits. Epigenome-wide association analysis was performed using the Illumina 27k chip.
Results: 51 of the 76 glycan traits studied have an additive genetic component (heritability, h (2) ) ≥ 0.5. In contrast, 12 glycan traits had a low genetic contribution (h(2)<0.35). We then tested for association between methylation levels and glycan levels (P<2 x10(-6)). Among glycan traits with low heritability probe cg08392591 maps to a CpG island 5' from the ANKRD11 gene, a p53 activator on chromosome 16. Probe cg26991199 maps to the SRSF10 gene involved in regulation of RNA splicing and particularly in regulation of splicing of mRNA precursors upon heat shock. Among those with high heritability we found cg13782134 (mapping to the NRN1L gene) and cg16029957 mapping near the QPCT gene to be array-wide significant. The proportion of array-wide epigenetic associations was significantly larger (P<0.005) among glycans with low heritability (42%) than in those with high heritability (6.2%).
Conclusions: Glycome analyses might provide a useful integration of genetic and non-genetic factors to further our understanding of the role of glycosylation in both normal physiology and disease.