Age-associated microRNA expression in human peripheral blood is associated with all-cause mortality and age-related traits

Recent studies provide evidence of correlations of DNA methylation and expression of protein-coding genes with human aging. The relations of microRNA expression with age and age-related clinical outcomes have not been characterized thoroughly. We explored associations of age with whole-blood microRNA expression in 5221 adults and identified 127 microRNAs that were differentially expressed by age at P < 3.3 × 10^-4 (Bonferroni-corrected). Most microRNAs were underexpressed in older individuals. Integrative analysis of microRNA and mRNA expression revealed changes in age-associated mRNA expression possibly driven by age-associated microRNAs in pathways that involve RNA processing, translation, and immune function. We fitted a linear model to predict 'microRNA age' that incorporated expression levels of 80 microRNAs. MicroRNA age correlated modestly with predicted age from DNA methylation (r = 0.3) and mRNA expression (r = 0.2), suggesting that microRNA age may complement mRNA and epigenetic age prediction models. We used the difference between microRNA age and chronological age as a biomarker of accelerated aging (Δage) and found that Δage was associated with all-cause mortality (hazards ratio 1.1 per year difference, P = 4.2 × 10^-5 adjusted for sex and chronological age). Additionally, Δage was associated with coronary heart disease, hypertension, blood pressure, and glucose levels. In conclusion, we constructed a microRNA age prediction model based on whole-blood microRNA expression profiling. Age-associated microRNAs and their targets have potential utility to detect accelerated aging and to predict risks for age-related diseases.