Deciphering the interplay between serum urate and bone mineral density: insights from observational and genetic...

Urate commonly influences bone metabolism due to its antioxidant properties, yet this relationship remains inadequately understood. Our study aims to investigate both the phenotypic and genetic relationships between urate and heel estimated bone mineral density (eBMD), a proxy marker for osteoporosis.

Leveraging UK Biobank individual-level data and summary statistics from the hitherto largest genome-wide association studies, we evaluated phenotypic associations using linear regression and restricted cubic splines (N = 424,105), then explored genetic correlations, causality, and pleiotropy through genome-wide cross-trait analyses (N_urate = 355,426; N_eBMD = 426,824).

Observational analysis suggested a significant linear association (β = 0.008, P = 3.35×10^-10) and an inverted U-shaped relationship (P_non-linear < 0.001) between urate and eBMD. Genome-wide genetic correlation was significant (r_g = 0.09, P = 2.01×10^-7), corroborated by six local signals at chromosomes 7, 9, 10, 14, 17, and 20. Cross-trait meta-analysis identified 237 pleiotropic loci, including 28 novel loci with five showing colocalization (PPH4 > 0.90). Gene-based analysis identified 278 shared genes. Transcriptome-wide association study revealed 150 shared genes, primarily enriched in tissues such as the tibial artery, fibroblasts, left ventricle, tibial nerve, and thyroid. One-sample Mendelian randomization demonstrated a causal effect of urate on eBMD (β = 0.038, P = 2.26×10^-21), as well as a non-linear causal relationship (P_{non-linearity} = 3.89×10^-3; P_{heterogeneity} = 0.02).

Our findings support phenotypic and genetic relationships between urate and eBMD, highlighting the etiological role of urate in osteoporosis and offering potential strategies for reducing osteoporosis risk.