Overview: Epigenetics refers to changes to the structure of DNA that affect patterns of gene expression without modifying the primary nucleotide sequence. Methylation of cytosine residues within CpG dinucleotides (5-methyl-cytosine) is one of several known epigenetic and a vast literature characterizes CpG sites and/or genomic regions that become either hypermethylated or hypomethylated with increasing age, suggesting a role for DNA methylation in biological aging.
Background: Recently, two groups independently published on two versions of the ‘epigenetic clock.’ (Horvath et al., 2013; Hannum et al., 2013). An epigenetic clock is an age predictor based on DNA methylation levels. Estimation is based on the linear combination of weighted methylation levels for a subset of CpGs across the genome. The specific CpGs used to define the respective epigenetic clocks were selected using elastic net regression aimed at producing the most accurate estimate of chronological age in DNA methylation (DNAm) data sets.
The Horvath epigenetic clock was developed using multiple publicly available DNAm datasets encompassing 51 different non-cancerous tissue and cell types. The model consists of 353 CpG sites, and its output of estimated epigenetic age is referred to as DNAm Age. Because this clock measure is derived from heterogeneous tissues, it is able to predict epigenetic age across many tissue and cell types (Horvath, 2013). Therefore, the Horvath clock is also known as pan tissue epigenetic clock. Hannum epigenetic clock is derived from DNAm profiling of whole blood in a cohort of 482 individuals of Caucasian or Hispanic ethnicity, aged 19 to 101. The model consists of 71 CpG sites that are together highly predictive of chronological age.
The term "clock" reflects the fact that epigenetic age estimates are highly correlated with chronological age: r>0.90 in individuals aged between 20 and 100 (as reviewed in Horvath and Raj 2018). A more recent clock, the skin & blood clock (Horvath et al., 2018) leads to superior accuracy in skin and blood cells including keratinocytes, buccal cells, fibroblasts, endothelial cells, blood, and saliva (Horvath et al., 2018). The original Horvath clock and the skin & blood clock (Horvath et al., 2018) are "life course" clocks because they apply to children and even prenatal samples (Horvath 2013, in Horvath and Raj 2018).
While the epigenetic clock was developed as an age predictor, it is generally the deviations of DNAm Age from chronological age that are of biological and clinical interest. When DNAm Age exceeds chronological age, the organism is said to experience epigenetic Age Acceleration (AgeAccel). AgeAccel is defined as the residual from regressing DNAm Age on chronological age, where a positive value indicates that epigenetic age is greater than expected. Unlike the raw DNAm Age estimated directly from methylation data, AgeAccel is uncorrelated with chronological age.
The Horvath group further characterizes AgeAccel in blood as either intrinsic or extrinsic. Intrinsic Epigenetic Age Acceleration (IEAA) refers to the residual resulting from regressing the Horvath estimate of epigenetic age on chronological age and measures of blood cell counts. IEAA thereby accounts for age-related changes in blood cell composition (decrease in naïve CD8+ T cells, increase in exhausted CD8+ T cells), and captures aspects of epigenetic aging that are preserved across cell and tissue types.
In contrast, Extrinsic Epigenetic Age Acceleration (EEAA) is derived from the Hannum estimate of epigenetic age by upweighting blood cell composition before regressing on chronological age. The Hannum estimate in its original form is already correlated with levels of certain types of blood cells, a consequence of its derivation from blood samples. By increasing the contribution of certain blood cell types whose levels are known to change with age, EEAA is able to capture aspects of immunosenescence. In Cox regression models of all-cause mortality, EEAA was a better predictor of death than the Hannum AgeAccel measure without upweighting (Chen et al., 2016).
More recently, researchers developed an alternative estimate of epigenetic age that was trained to predict ‘Phenotypic Age’, rather than chronological age. Phenotypic age is calculated using clinical measures, selected to predict aging-related mortality. Therefore, this new epigenetic clock is known as DNAm PhenoAge, or Levine DNAmAge. The DNAm PhenoAge measure was derived in two steps. First, this research group identified ten clinical biomarkers (including chronological age) that were predictive of mortality in the NHANES III study (N=9926, 23-year follow-up). These ten variables were used to define a phenotypic age estimate. In the second step, phenotypic age was related to DNAm levels in blood. The process yielded 513 CpG sites predictive of phenotypic age, and the linear combination of these CpGs defines DNAm PhenoAge. The new measure was tested against its predecessors in its ability to predict mortality and morbidity in five independent large cohort samples, where DNAm PhenoAge was found to outperform the epigenetic age measures from Horvath and Hannum (Levine et al., 2018).
Authors and Reviewers
This summary was written by Katerina Protsenko and Aric A. Prather, PhD, and reviewed by Steven Horvath, PhD, ScD & Morgan E. Levine, PhD.
Hannum, G., Guinney, J., Zhao, L., Zhang, L., Hughes, G., Sadda, S., Klotzle, B., Bibikova, M., Fan, J.B., Gao, Y., Deconde, R., Chen, M., Rajapakse, I., Friend, S., Ideker, T., Zhang, K., 2013. Genome-wide methylation profiles reveal quantitative views of human aging rates. Mol Cell 49, 359-367.
Horvath, S., 2013. DNA methylation age of human tissues and cell types. Genome biology 14, R115.
Horvath, S., Raj, K., 2018. DNA methylation-based biomarkers and the epigenetic clock theory of aging. Nature Review Genetics 19, 371-384.
Horvath, S, Oshima, J., Martin, G.M., Lu, A.T., Quach, A., Cohen, H., Felton, S., Matsuyama, M., Lowe, D., Kabacik, S., Wilson, J.G., Reiner, A.P., Maierhofer, A., Flunkert, J., Aviv, A., Hou, L., Baccarelli, A.A., Li, Y., Stewart, J.D., Whitsel, E.A., Ferrucci, L., Matsuyama, S., Raj, K., 2018. Epigenetic clock for skin and blood cells applied to Hutchinson Gilford Progeria Syndrome and ex vivo studies. Aging 10, 1758-1775.
Levine, M.E., Lu, A.T., Quach, A., Chen, B.H., Assimes, T.L., Bandinelli, S., Hou, L., Baccarelli, A.A., Stewart, J.D., Li, Y., Whitsel, E.A., Wilson, J.G., Reiner, A.P., Aviv, A., Lohman, K., Liu, Y., Ferrucci, L., Horvath, S., 2018. An epigenetic biomarker of aging for lifespan and healthspan. Aging 10, 573-591.