Solving the mysteries of bioscience
Foundational Science Fuels Breakthroughs
Inspiring Next-Generation Scientists
‘Epigenetic clocks’ vary with disease and even by organ, but scientists are still trying to understand why
10 min read
By Leila Okahata
Age is more than just the number of candles on a birthday cake. In fact, the number of birthdays you’ve had may not be the most accurate way to determine how old your body is.
Instead, scientists are looking to measure biological age. Whereas chronological age is the number of years since you were born, biological age is the age of your body based on your physical health status. With biological factors considered, biological age may not only be a better predictor of how old an individual is — what if you are actually older or younger than you think you are? — but it also may be a better predictor of lifespan and health.
There are many ways to calculate biological age, such as through glucose levels, cholesterol levels, and even walking speed. But over the last decade, scientists have found that DNA has wrinkle patterns of its own: the epigenome.
Think of the epigenome as accessories that our DNA wears or takes off to turn genes on and off. These epigenetic modifications are critical for cell differentiation and are responsible for the subtle differences between identical twins and for the spots on calico cats.
Steve Horvath, Ph.D., a UCLA aging researcher and an Allen Distinguished Investigator, studies a form of epigenetic modification known as methylation, in which methyl groups — small chemical tags — are attached to DNA. As we age, our methylation patterns change in predictable ways: Certain DNA regions lose methylation, while others gain methylation. By measuring and averaging those patterns, Horvath created in 2013 the first test that could predict epigenetic age (an aspect of biological age) across multiple human tissues — a test he dubbed the epigenetic clock.
Using his epigenetic clock on 8,000 healthy samples of 51 different organs and cell types, Horvath found that the epigenetic age of most tissue samples matched the chronological age of the donor. But some organs differed noticeably from their chronological age. Female breast tissue, for example, was two to three years older than the rest of the body. What did this mean?
To Horvath’s surprise, his study was hinting that the body does not age in unison — different parts of the body age faster than others.
Every tissue and organ in the body has its own age. In fact, the youngest part of the human body is the part of the brain called the cerebellum (up to 15 years younger than your chronological age), whereas the oldest is female breast tissue (up to three years older), Horvath said.
But if we are a mosaic of organ ages, how old are we truly? How can our age encompass all the various ticking clocks in our bodies? Horvath too ponders these questions.
“In theory, each of your organs, each of your tissues, has a different age. Which comes back to the discussion: What’s really the meaning of biological age?” Horvath said. “Because imagine, your kidney is 30 years old, your liver is 35, your blood is 25, your brain is 18. You have all these different ages, and how do you summarize them in one overall number?”
Scientists also don’t understand why the clock ticks faster in certain tissues. Horvath speculates that, regarding the youthful cerebellum, it could be because neurons don’t divide. Or maybe the brain has added age-protection features. For rapidly maturing breast tissue, maybe it is because hormones proliferate the cells. Nevertheless, the mechanism, or rather the clockwork, behind aging is still unknown.
However, scientists have uncovered some of its gears. Our environment can influence our epigenetic modifications. This includes exposure to pollutants; lifestyle factors like diet, smoking and alcohol consumption, and exercise; but, most notably, infection.
Scientists have confirmed one infection that accelerates age: HIV.
In a recently published study by UCLA researchers and colleagues, scientists used epigenetic clocks to analyze blood samples from people living with HIV. Within three years of initial infection, these individuals showed significant age acceleration from their chronological age, ranging from 1.9 to 4.8 years older.
Health conditions like HIV can exacerbate age, which is a problem not only in terms of lifespan but also healthspan — the number of years lived disease-free. According to Beth D. Jamieson, Ph.D., senior author of the study and an HIV researcher at UCLA, accelerated aging may put individuals at greater risk for age-related morbidities like diabetes, cognitive decline, frailty, and cardiovascular disease. Even those treated with drugs to suppress HIV infection, called antiretroviral therapy, may still be at risk.
“We began hearing from clinicians that people who were on antiretroviral therapy and were successfully treated — meaning their viral loads had gone down to undetectable — were coming into the clinic looking like and having many of the comorbidities of aging that you would expect from somebody who was maybe 10 years older than they were,” Jamieson said.
HIV has a deep impact on the epigenome that even two years of antiretroviral therapy doesn’t fully reverse, but not all viral infections accelerate age. In fact, not all health conditions or diseases have a direct causal relationship with accelerated aging like HIV. Obesity is associated with age acceleration in the liver; Down syndrome is associated with age acceleration in the brain and blood; hypertension is associated with age acceleration in the heart, kidney, liver, and muscle. But as of today, these are only associations, not causations. Does cancer cause accelerated aging, or does accelerated aging cause cancer? Or is there instead a third factor, such as smoking, that triggers both accelerated aging and cancer?
“Unfortunately, that’s the situation with quite a lot of diseases, where we need to disentangle the cause and effect,” Horvath said.
But that uncertainty is also what drives Horvath and other scientists to find the answers.
As more scientists seek to find the underpinnings of accelerated aging, just as many seek to discover the agents that rewind or slow the ticking.
In 2019, Horvath collaborated with a start-up company called Intervene Immune, led by biologist Greg Fahy, Ph.D., in a small clinical trial to rejuvenate the thymus, the organ above the heart that produces immune cells. The study included nine healthy men aged 51 to 65. Treating the participants with a growth hormone and two diabetes medications, the scientists found that the drugs reversed the epigenetic age of the thymus by 2.5 years after a year of treatment.
But is 2.5 years a lot? Does being 49 instead of 51 change much?
“It’s a huge effect. The reason is because, fundamentally, you stopped the age,” said Horvath, who is also a principal investigator at the anti-aging startup Altos Labs. “You could say, ‘Well, you didn’t cut the age by half.’ But you did stop the aging, which is good. If you could stop my aging right now, I would be very happy.”
However, the scientists only followed the participants for six months after the treatment, so whether the reversed age persists for longer or if lifespan is lengthened is unknown. Additionally, the study was limited to a small group of nine individuals. The scientists are currently performing a replication study with a larger group of 85 participants, hoping to find the same results.
There could also be a link between HIV treatment and age reversal. In Jamieson’s 2020 study on antiretroviral therapy, she and her colleagues found a small reversal in the epigenetic age of people living with HIV 18 to 24 months after starting antiretroviral therapy. Although biological age was not fully reset to what it was before HIV infection — which may or may not explain why these individuals still develop age-related conditions and diseases — Jamieson is curious about the future potential of this therapy.
“Is this change permanent? We know it can be modified at least slightly by antiretroviral therapy, but what happens with long-term antiretroviral therapy?” she said.
Jamieson and colleagues are now studying the long-term epigenetic impacts of antiretroviral therapy, checking on individuals up to 17 years later after starting treatment. But she emphasized scientists might learn more about age reversal by studying individuals who are epigenetically younger than their chronological age.
“What about these individuals who are at the other end of the spectrum? Is there something they’re doing that we can mimic and everybody else can apply to increase their health and minimize the bad outcomes? That’s a whole other area to explore,” she said.
Imagine one day you enter the doctor’s office for your annual health checkup. It’s the usual routine: Your physician takes your height and weight, followed by questions about your diet, exercise habits, and smoking and alcohol consumption. She types these values into the computer and takes a blood sample. Here’s what’s new: All of this is sent to an epigenetic clock to be analyzed, and it spits out your predictive risk for a plethora of health conditions and diseases: cardiovascular disease, diabetes, certain cancers, and more. These are the clinical applications of epigenetic clocks that Riccardo Marioni, Ph.D., a genetic epidemiology researcher at the University of Edinburgh, hopes to achieve.
Marioni is searching for the epigenetic signatures of basic physical traits, like BMI and sex, and lifestyle behaviors, like smoking and alcohol habits. Discovering the marks, or biomarkers, that these traits leave on our epigenome can potentially improve predictions of disease risk, he said. Moreover, biomarkers can serve as a more objective record of one’s behaviors, leading to greater accuracy of risk predictions, he added.
“When a physician asks, ‘Are you a smoker? How much do you smoke?’ and similarly with alcohol consumption, ‘How much do you drink?’, it’s reliant on self-report. There’s a degree of subjectivity,” he said. “Biomarkers should hopefully be more objective, less prone to recall bias.”
However, we are far from epigenetic clocks as clinical tools. These tests are expensive and still have large errors associated with their predictive capabilities. Scientists are also unsure how much epigenetic age tracks with overall biological age. Horvath said that the best way to explain this uncertainty is through the story of The Blind Men and The Elephant: a group of blind men touch an elephant for the first time, each feeling different parts. Some touch the trunk, some the tails, some the legs. The blind men then discuss what they think the elephant looks like.
“Now I certainly touch a part of the elephant with DNA methylation,” Horvath said. “The question is, how large a part? Do I only touch a part of the foot? Or do I actually cover a lot of the elephant? That’s a bit of a debate.”
Nevertheless, epigenetic clocks are one of the most accurate measures of biological age available today. And if there is one thing these clocks prove, it is that the human population is growing older and living exceptionally longer than we ever had in the past.
But now, it is time to figure out how to make that long life lived better.
“It’s not just about increasing lifespan, but also increasing healthspan, so that you stay healthier for longer rather than live longer with more chronic ailments,” Marioni said.
The Allen Distinguished Investigator awards are funded by the Paul G. Allen Family Foundation. The Paul G. Allen Frontiers Group, a division of the Allen Institute, recommends funding and supports the administration of the awards.
Leila Okahata is a former Editorial Intern in the Communications department at the Allen Institute. She is a science writer focusing on life science research and hopes to communicate complex scientific information into deeply compelling stories. She is a Science and Health news reporter for the Daily Bruin, UCLA’s student newspaper. She is currently a fourth-year undergraduate at UCLA majoring in microbiology, immunology, and molecular genetics with a minor in professional writing.
Get in touch at firstname.lastname@example.org.