Patterns of MicroRNA Expression as a Biomarker of Aging
As a complement to the DNA methylation based biomarkers of aging, researchers are also finding that patterns of microRNA levels can serve a similar purpose. These tools offer the potential for a rapid assessment of candidate rejuvenation therapies rather than having to run lengthy life span studies, something that is prohibitively expensive for most research groups even in mice, and entirely out of the question in humans. It is hoped that a generally agreed upon biomarker of aging, a low-cost test that fairly accurately reflects biological age, defined as the burden of cell and tissue damage that causes dysfunction and death, will speed up progress towards the development of rejuvenation treatments. The advent of senescent cell clearance as a viable rejuvenation therapy should greatly help the development and validation of such biomarkers: the two lines of development will support one another.
Human aging is a complex process that has been linked to dysregulation of numerous cellular and molecular processes. Recent studies have revealed that human aging can be characterized by changing patterns of DNA methylation and expression of protein-coding genes. A growing body of research suggests that aging is associated with changes in DNA methylation both genome-wide and at specific C-G dinucleotide (CpG) loci. At the messenger RNA (mRNA) level, a recent meta-analysis of whole-blood gene expression in ~15,000 individuals identified 1497 mRNAs that are differentially expressed in relation to age. An age predictor based on mRNA expression (i.e., mRNA age) highlighted genes involved in mitochondrial, metabolic, and immune function-related pathways as key components of aging processes. The difference between mRNA age and chronological age correlated with many metabolic risk factors including blood pressure, total cholesterol levels, fasting glucose, and body mass index (BMI).
MicroRNAs (miRNAs) are a class of small noncoding RNAs that downregulate protein-coding genes by either cleaving target mRNAs or suppressing translation of mRNAs into proteins. Research in a Caenorhabditis elegans model system revealed changes in miRNA expression in relation to lifespan and longevity. In humans, highly specific miRNA expression patterns are correlated with many age-related diseases including cardiovascular disease and cancer. Recent studies have examined differentially expressed miRNAs in relation to age in whole blood, peripheral blood mononuclear cells (PBMC), and serum. These studies, however, were based on small sample sizes, limiting the power to investigate age-related changes in miRNA expression. We hypothesized, a priori, that it would be possible to create a miRNA signature of age that is predictive of chronological age and that age prediction based on miRNA expression is biologically meaningful and can be used as a biomarker of risk for age-related outcomes including all-cause mortality.
In a previous study, we measured miRNA expression in whole blood from more than 5,000 Framingham Heart Study (FHS) participants. We investigated the heritability of miRNA expression and performed a genome-wide association study (GWAS) of miRNA expression. Our results showed that miRNAs are under strong genetic control. In the present study, we further investigated whole-blood miRNA expression in relation to chronological age in FHS participants. We identified 127 miRNAs that were differentially expressed in relation to chronological age, and performed internal validation by splitting the samples 1:1 into two independent sample sets. An integrative miRNA-mRNA coexpression analysis and miRNA target prediction revealed many age-related pathways underlying age-associated molecular changes. We also defined and evaluated an age predictor based on miRNA expression levels (i.e., miRNA age). Our results indicate that the difference between miRNA age and chronological age is associated with multiple age-related clinical traits including all-cause mortality, coronary heart disease (CHD), hypertension, blood pressure, and glucose levels.