Association of serum 25-hydroxyvitamin D levels, vitamin D-binding protein levels, and diabetes mellitus: Two-sample Mendelian randomization

Abstract

Studies have suggested that Vitamin D deficiency is associated with the occurrence of both type 1 and type 2 diabetes, and that vitamin D-binding proteins (VDBP) are necessary for metabolic stress in pancreatic α-cells. However, the causal relationship between serum 25-hydroxyvitamin D [25(OH)D] levels, VDBP, and the risk of diabetes mellitus (DM) remains unclear. Mendelian randomization (MR) was used to investigate the causal relationship between 25(OH)D, VDBP, and DM. Relevant recent data were downloaded from the NHGRI-EBI Catalog of published genome-wide association studies (GWAS) and filtered for single nucleotide polymorphisms (SNPs). We used multiple MR methods, including inverse variance weighting (IVW), and performed sensitivity analyses to detect whether pleiotropy or heterogeneity biased the results. There was a causal relationship between genetically predicted VDBP levels and serum 25(OH)D levels, and serum 25(OH)D levels increased with increasing VDBP levels (IVW: β = 0.111, OR = 1.117, 95% CI:1.076-1.162, P = 1.41 × 10-8). There was no causal relationship between the genetically predicted VDBP levels, serum 25(OH)D levels, and DM (VDBP: IVW β:0.001, OR:1.001, 95% CI:0.998-1.003, P > .05; 25(OH)D: IVW β: -0.009, OR:0.991, 95% CI:0.982-1.001, P = .068). Sensitivity analysis indicated that horizontal pleiotropy was unlikely to bias causality in this study. MR analysis results demonstrated a positive causal relationship between VDBP levels and serum 25(OH)D levels in the European population. The 25(OH)D and VDBP levels were not causally related to an increased risk of diabetes.