Reproducibility Of Gene Expression And DNA Methylation Biomarkers For Aging In Peripheral Whole Blood

The main feature of aging is degradation of physiological function which has individual difference. Instead of chronological age, seeking for reliable and available biomarkers to evaluate the aging state has important significance for pathogenesis research and auxiliary diagnosis of aging-related diseases such as Alzheimer’s disease(AD). Using peripheral blood which is easy to measure clinically, more and more studies have found some gene expression and DNA methylation biomarkers for aging. However, no studies have systematicly evaluated the reproducibility of aging signatures at present.Based on multiple datasets for methylation profiles and expression profiles of peripheral whole blood, we analysed whether age-associated differentially expressed genes(aDEG) and methylated sites(aDMS) obtained from independent datasets were consistent respectively. The results showed that aDEG and aDMS between different datasets are highly consistent(p<4.72*E-05), which indicates reproducibility of aging-related gene expression and DNA methylation biomarkers which obtained by high-throughput technology in blood. We further estimated the percentages of main leukocyte subtypes in every sample by using deconvolution algorithm and analyzed age-related molecular changes on leukocyte subtypes. The results showed:(1) with aging, the proportion of myeloid cells in whole blood increases significantly and the proportion of lymphoid cells decreases significantly;(2) aDEG observed in whole blood mainly reflect differentially expression signals caused by change of blood components with aging;(3) aDMS in whole blood is dominated by cell-intrinsic DNA methylation changes. Finally, we compared a DEG and aDMS found in normal blood with corresponding signatures of AD, respectively. The results indicated:(1) aDMS over 65 years of age, in contrast to the entire age group, have a higher consistency with AD, including some genes which have been reported as potential biomarkers for AD;(2) there’s no significant correlation between aDEG on blood cells and differentially expressed genes related to AD.In summary, our study confirmed that the age-associated gene expression and methylation biomarkers in whole blood are reproducible. With age increases, the proportions of blood cell subsets can change and have a significant impact on gene expression biomarkers in whole blood, with a relatively smaller impact on DNA methylation biomarkers. aDMS over 65 years old in peripheral whole blood may be potential molecular markers of alzheimer’s disease, which deserve further study.