Mendelian Randomization Revealed Potential of mTOR Inhibitors for Treatment of Osteoporosis: Evidence From GWAS and Transcriptome Data.

Aims: Numerous preclinical studies suggested that targeted inhibition of mammalian target of rapamycin (mTOR) may be beneficial for the treatment of osteoporosis. However, the relevance of these findings to human populations remains unclear. We hypothesized that lifelong genetically higher mTOR expression leads to reduced bone mineral density (BMD), and we use Mendelian randomization to estimate its causal effect. Background: We aimed to utilize existing genome-wide association studies (GWAS) and expression quantitative trait locus (eQTL) data to comprehensively evaluate whether genetically proxied mTOR expression is associated with BMD using the Mendelian randomization method. Methods: Independent cis-eQTLs for the gene-encoding mTOR were selected as instrumental variables (IVs). Two large GWAS meta-analysis cohorts were used for BMD outcomes. Summary data-based Mendelian randomization (SMR), inverse variance-weighted Mendelian randomization (IVW-MR), and multivariable Mendelian randomization (MVMR) were applied to estimate the association between genetically proxied mTOR expression and BMD. Phenome-wide association study (PheWAS) analysis was conducted to explore potential horizontal pleiotropy and genetically associated phenotypes related to mTOR. Bayesian colocalization was performed to assess whether mTOR expression and BMD shared a common causal single-nucleotide variant (SNV). Blood transcriptome data were analyzed to further evaluate expression patterns of mTOR in relation to BMD status. In addition, molecular docking was conducted to explore potential interactions between mTOR and representative fatty acids. Analyses followed STROBE-MR guidance. Results: SMR analysis indicated that increased genetically proxied mTOR expression in blood was associated with reduced BMD (OR = 0.8130-0.9340), except in individuals younger than 30 years. Colocalization analysis suggested that mTOR expression and BMD may share a common genetic locus (posterior probability for H4 = 77.7%). PheWAS identified an association between mTOR and basal metabolic rate (Beta = -16.769, FDR = 2.21E - 25), indicating potential metabolic implications. Further analyses evaluated associations between mTOR-related genetic variation, circulating fatty acids, hormone levels, and BMD. MVMR results suggested that in individuals aged 30-45 years, the association between genetically proxied mTOR expression and BMD may be influenced by blood fatty acid levels (OR = 0.8823-0.9020). In older age groups (45-60 years and ≥ 60 years), genetically predicted insulin levels were associated with attenuation of these metabolic effects. Blood transcriptome analysis showed no significant difference in mTOR expression between high- and low-BMD groups, indicating that peripheral transcript levels were not detectably associated with BMD status. Arachidonic acid demonstrated the highest predicted binding affinity to mTOR (-5.9 kcal/mol), followed by palmitic acid (-5.0 kcal/mol), oleic acid (-4.9 kcal/mol), linoleic acid (-4.8 kcal/mol), and stearic acid (-4.4 kcal/mol). Given the modest binding energies and lack of experimental validation, these findings should be interpreted cautiously and are intended to provide preliminary mechanistic hypotheses rather than confirm functional interactions. Conclusions: Genetic evidence indicates that mTOR-related pathways may contribute to variation in BMD, with age-dependent metabolic associations. The molecular docking results provide exploratory structural hypotheses that may inform future mechanistic studies but do not constitute functional or translational validation. Although these findings do not establish the clinical efficacy of pharmacological mTOR inhibition, they offer human genetic evidence supporting further experimental and pharmacological investigation of mTOR as a potential therapeutic target in osteoporosis.