Assessing the Associations of Blood Metabolites With Osteoporosis: A Mendelian Randomization Study

Context: Osteoporosis is a metabolic bone disease. The effect of blood metabolites on the development of osteoporosis remains elusive. Objective: To explore the relationship between blood metabolites and osteoporosis. Design and Methods: We used 2286 unrelated white subjects for the discovery samples and 3143 unrelated white subjects from the Framingham Heart Study (FHS) for the replication samples. The bone mineral density (BMD) was measured using dual-energy X-ray absorptiometry. Genome-wide single nucleotide polymorphism (SNP) genotyping was performed using Affymetrix Human SNP Array 6.0 (for discovery samples) and Affymetrix SNP 500K and 50K array (for FHS replication samples). The SNP sets significantly associated with blood metabolites were obtained from a reported whole-genome sequencing study. For each subject, the genetic risk score of the metabolite was calculated from the genotype data of the metabolite-associated SNP sets. Pearson correlation analysis was conducted to evaluate the potential effect of blood metabolites on the variations in bone phenotypes; 10,000 permutations were conducted to calculate the empirical P value and false discovery rate. Results: We analyzed 481 blood metabolites. We identified multiple blood metabolites associated with hip BMD, such as 1,5-anhydroglucitol (P-discovery < 0.0001; P-replication = 0.0361), inosine (P-discovery = discovery = 0.0018; P-replication = 0.0256), theophylline (P-discovery = 0.0048; P-replication = 0.0433, gamma-glutamyl methionine (P-discovery = 0.0047; P-replication = 0.0471), 1-linoleoyl-2-arachidonoyl-GPC (18:2/20:4n6; P-discovery = 0.0018; P-replication = 0.0390), and X-12127 (P-discovery = 0.0002; P-replication = 0.0249). Conclusions: Our results suggest a modest effect of blood metabolites on the variations of BMD and identified several candidate blood metabolites for osteoporosis.