top of page
rusty dna.png

Tick-Tock: The Story of our Epigenetic Clocks

Magdalena Drożdż


Scientists and philosophers have been trying to solve the secrets of eternal life. The Epic of Gilgamesh tells a story of an ancient Sumerian king desperately trying to find a way to stay immortal. Needless to say he failed, but his efforts inspired a group of German teenagers in the 1980s. Steve Horvath, his twin brother Markus and their friend Jörg Zimmerman discussed all things maths, physics and philosophy. As the meetings often led to discussions about human lifespan, the trio named them ‘the Gilgamesh Project’. At the final meeting in 1989, they decided to devote their careers to the study of ageing and prolonging a healthy human lifespan.


At the time, ageing research was going through a series of ground-breaking discoveries. A year earlier, Robert Moyzis published the sequence of human telomeres, which protect the ends of our chromosomes from damage upon cell divisions. As the cells divide, telomeres shorten, eventually reaching a critical limit. Soon it was confirmed that the length of the telomeres can indicate the relative age of the observed cells. Having a marker allowing to assess the cell age opened the doors for more research on ageing and longevity, inspiring a multitude of studies. At the same time, the field of epigenetics was also emerging. It helped us appreciate the regulatory mechanisms that beyond the level of the DNA sequence, affecting gene expression by causing chemical or structural changes. One of the most studied epigenetic modifications  is cytosine methylation, where methyl (-CH3) groups are added to cytosines followed by guanines (CpG sites). Of the 28 million of CpGs, over 60% are methylated - a vital process for correct gene regulation and cell development. A true breakthrough in methylation studies came with the development of high-throughput chips allowing for simultaneous examination of the methylation stage of thousands of CpGs in the human genome.



Twenty years after the Gilgamesh project started Steve Horvath, equipped with a PhD in mathematics and ScD in biostatistics, began studying the correlation between methylation patterns and sexual orientation. While these initial studies did not yield any results, Horvath then tested the data against the sample donors’ age. To his surprise, he found that methylation at just a few CpG sites could predict the donor’s age.  After months of collecting enormous pan-tissue datasets, improving the algorithm, and facing rejections from sceptical reviewers, Horvath published the first age estimator in 2013. The paper sparked excitement among numerous researchers who used the model to accurately predict the age of their own samples, validating the model’s success. 


His first age estimator, Horvath’s clock, is based on the methylation status of 353 CpGs scattered across the genome. These sites get methylated or unmethylated in life, acting as stable marks on DNA, like rust on metal. Knowing the status of these positions, Horvath’s clock predicts the biological age of any tissue in the organism, far more accurately than the telomere model. Several other estimators have been published to date such as Hannum’s clocks which use the age of leukocytes to predict age. The epigenetic clocks indicate tissues’ biological age. This can be higher or lower than chronological age starting at birth. Biological or epigenetic clocks indicate the health and longevity of tissues and entire organisms, and people with lower epigenetic age tend to live longer than their calendar age-matched peers. In 2015, a study looked into a group of Italian centenarians and found their epigenetic age was about 9 years younger than their chronological age. This effect was also observed in their offspring when compared with age-matched controls. The development of the epigenetic clocks finally provided a reliable ageing biomarker, paving the way to a new generation of longevity research. Description of the studies is beyond the scope of this article, but many come down to a simple conclusion: your grandma was right – eat your veggies and exercise regularly.



As we know, the calendar age does not always reflect our health. The discovery of DNA methylation markers correlated with a set of clinical measures, plasma proteins and smoking pack-years led to the development of two other clocks: PhenoAge and GrimAge, which are strongly associated with the general health of the individual. The discrepancies between GrimAge and calendar age indicate acceleration or deceleration of ageing processes and are associated with an array of clinical phenotypes such as walking speed or sustained attention reaction time (Lu et al., 2019; McCrory et al., 2021). An analysis of the Women’s Health Initiative data showed that accelerated epigenetic ageing could predict the future onset of lung cancer (Levine et al., 2015). Other studies found connections to cardiovascular health, time-to-cancer or Parkinson’s disease, presenting epigenetic ageing as predictive biomarkers (Horvath et al. 2015, Lu et al., 2019; Joyce et al., 2021).



These measurements have a long way to enter the clinic but represent the improvement of markers that promise to advance precision medicine. Unfortunately, we still lack a deeper understanding of what makes them “tick”. The field faces a chicken and egg question: are the epigenetic changes the reason for tissue ageing, or is there another process that results in methylation changes, making the biomarkers just a clock face? This presents a challenge for the lifespan-elongation studies, but should not be an obstacle for developing useful clinical biomarkers. One is sure for certain – the ageing research is blooming, and many questions will be answered in the coming years.




Horvath, S. and Ritz, B.R. (2015) ‘Increased epigenetic age and granulocyte counts in the blood of Parkinson’s disease patients’, Aging, 7(12), pp. 1130–1142. doi:10.18632/aging.100859.

Joyce, B.T. et al. (2021) ‘Epigenetic Age Acceleration Reflects Long-Term Cardiovascular Health’, Circulation Research, 129(8), pp. 770–781. doi:10.1161/CIRCRESAHA.121.318965.

Lu, A.T. et al. (2019) ‘DNA methylation GrimAge strongly predicts lifespan and healthspan’, Aging, 11(2), pp. 303–327. doi:10.18632/aging.101684.

McCrory, C. et al. (2021) ‘GrimAge Outperforms Other Epigenetic Clocks in the Prediction of Age-Related Clinical Phenotypes and All-Cause Mortality’, The Journals of Gerontology: Series A, 76(5), pp. 741–749. doi:10.1093/gerona/glaa286.

Quach, A. et al. (2017) ‘Epigenetic clock analysis of diet, exercise, education, and lifestyle factors’, Aging, 9(2), pp. 419–446. doi:10.18632/aging.101168.

bottom of page