Prospective evidence for the gut–bone axis in osteoporotic fractures: Insights from genetic prediction and metabolite mediators

Abstract

Objective

To clarify the causal role of gut microbiota in osteoporotic fracture and identify underlying metabolic pathways.

Method

We constructed genetic risk scores (GRS) for 211 microbial taxa using summary statistics from the MiBioGen consortium, and evaluated their associations with incident osteoporotic fracture in 446,927 participants of European ancestry in the UK Biobank. Osteoporotic fracture was defined by a composite of low bone mineral density (T-score ≤ −1.0 at any site) and clinically confirmed fracture events. Cox proportional hazards models were used for time-to-event analysis, with various covariate adjustment and sex-stratified evaluations. Genome-wide association analysis was performed to identify osteoporotic fracture-related loci. We then (i) performed two-sample Mendelian randomization (MR) with independent outcome GWASs (UK Biobank & FinnGen R12) and (ii) rebuilt GRSs with Dutch Microbiome Project (DMP) instruments for external replication. Mediation analysis explored HDL-related metabolic traits as potential biological intermediates.

Results

We identified several microbial GRSs nominally significant associated with osteoporotic fracture risk. Notably, genetically predicted Eubacterium xylanophilum group and Enterobacteriaceae showed consistent associations with increased osteoporotic fracture risk across fully adjusted models (HR = 1.426, P = 0.023), while taxa such as Sellimonas, Ruminococcus2, and Escherichia/Shigella were protective. These associations were stronger in females. GWAS identified one genome-wide significant locus on chromosome 18 (rs146540715) for osteoporotic fracture. Two-sample MR analyses provided suggestive evidence for a potential causal relationship between Eubacterium xylanophilum group and osteoporotic fracture, and directionally consistent associations for Enterobacteriaceae across UKBB and FinnGen datasets. No evidence of pleiotropy was detected. Mediation analysis revealed that Eubacterium xylanophilum group may influence osteoporotic fracture via HDL metabolism, particularly through cholesteryl esters in HDL, explaining 3.34 % of the total effect.

Conclusions

Our results provide exploratory genetic and epidemiological support for a link between specific gut microbes and osteoporotic fracture risk (potentially acting in part through HDL-related pathways). Larger independent studies are needed to confirm these associations before any clinical translation is considered.