A patient sits in the exam room, waiting to be seen by his doctor. His mind drifts, as it often does, to his family. He pictures his granddaughter’s face – the freckles, the missing teeth, the bright smile. But when she lifts her hand to wave, it vanishes just as it leaves the center of his vision. He turns his head to find it, like searching with a flashlight whose beam has narrowed to a small circle. The living room where he sits is all missing pieces: no chair legs, a missing lamp, a door without a handle. Everything at once is both there and not there, present but just out of frame. He no longer trusts doorframes, curbs, or crowded sidewalks. The danger is not where he looks, it’s where he can’t. This is the grim reality of glaucoma. And it only gets worse.
Back in the room, the doctor arrives. She wants to talk about a new study the clinic is joining to treat patients with glaucoma, a first-in-human trial of a gene therapy called ER-100. The treatment, she says, would deliver three genes into his optic nerve, aiming to coax his cells to behave more like they did when he was young. She explains that the origins of this treatment are not related to glaucoma or even the eye at all, but basic research into why we age. This research has suggested that as we get older, cells can begin to “forget” their identity. In the eye, that means cells gradually lose the precise functions required to carry visual information to the brain. Some scientists now view diseases like glaucoma through this lens, less as a specific problem of the eye and more as what happens when cells in the optic nerve lose their identity. And in fact, this principle appears to apply more and more broadly across tissues and diseases.
What if aging and age-related diseases – Alzheimer’s disease, Type 2 diabetes, heart disease – are ultimately just manifestations of cells forgetting how to be young?
Aging, reimagined as an information problem
ER-100 represents the first clinical test of a larger idea about aging itself, an idea called the Information Theory of Aging (ITOA). For ER-100 to make any sense, you have to accept a strange premise: cells carry memories. Cells remember who they should be, which genes to turn on and off, how to organize themselves into a tissue. These memories are written into patterns of gene expression, managed by a control system known as the epigenome. In glaucoma, as in many age-related diseases, those memories start to blur. Gene expression patterns become dysregulated, cellular identity weakens, and function declines.
The ITOA invites us to shift perspective. Rather than viewing the body simply as a collection of mechanical parts that wear out over time, we can see it as a dynamic system for storing, preserving, and interpreting biological information. Aging, in this framework, is the progressive loss of the cell’s ability to read its own instructions correctly, and therefore to maintain its identity.
Aging biology is crowded with theories, but none has yet earned the status of a single, agreed-upon explanation. Most historical theories of aging are sorted into two broad categories: error-based and program-based. Debates in geroscience have long been framed as “damage vs. program.” For much of the twentieth century, aging research was dominated by damage-accumulation theories: oxidative stress, free radicals, telomere shortening, and the promise of antioxidants shaped both the scientific conversation and, later, the consumer marketplace. The central idea was straightforward: cells accumulate molecular damage over time, and that damage eventually overwhelms repair systems. Yet the evidence we have – from comparative lifespans, longevity mutations, and drugs that extend life – can almost always be interpreted in both directions. The same intervention may look like “less damage” in one framework and “turning down a quasi-programmed process” in another.1-3
Only more recently have theorists started to talk about aging as a problem of biological information – a gradual loss of the precise patterns of gene activity and communication that keep cells coordinated. ITOA grows out of this shift, asking not just how much damage there is, but what it does to the instructions that tell cells how to use their genes. Work from laboratories led by Dr. Leonard Guarente (Massachusetts Institute of Technology), Dr. Brian Kennedy (National University of Singapore), Dr. Shin-ichiro Imai (Washington University), and Dr. David Sinclair (Harvard Medical School) focused on chromatin regulation and the epigenome as central drivers of aging. Their findings point to a consistent story: changes in gene silencing and epigenetic regulation might be driving aging rather than simply reflecting it. This story was reinforced by later work in mammals showing age-related epigenetic alterations, shifts in chromatin structure, and DNA methylation patterns that reliably tracked with biological age. In other words, aging might have less to do with damage to the genome itself, and more to do with gradual errors in how that DNA is read.
Our genomes are the closest thing biology has to digital hardware. DNA is a long string of base pairs copied with remarkable accuracy from cell to cell and across generations. Your DNA sequence is essentially the same in a skin cell as in a neuron, in childhood and in old age. Yet those cells behave very differently, and they change in different ways over time. The difference lies not in the letters of the code, but in how that code is read.
By changing where chemical tags are added to DNA and the proteins that spool DNA, the epigenome decides which genes are switched on or off. Unlike the genome, it is largely dynamic. The epigenome is constantly being nudged by signals from inside and outside the cell – stress, diet, inflammation, DNA damage, and time. That flexibility is what lets cells adapt, but it also makes the system fragile. Over decades, epigenetic patterns can drift, tags end up in the wrong places, and the instructions are eventually so badly eroded that the cell forgets how to function.
From the perspective of the ITOA, genetic information itself is comparatively durable. You can even read the DNA of Neanderthals, a reminder that Jurassic Park’s central idea (reading ancient DNA) has a real scientific basis, even if resurrecting dinosaurs does not. What fails with age is our cells’ ability to read and interpret that information correctly. The theory proposes that aging is driven by the gradual loss and corruption of epigenetic information that tells cells how to use the genome, rather than by DNA mutations or other purely physical damage alone. Aging therefore is what happens when these information-maintenance systems can no longer keep gene expression programs on track, and cells become less coordinated, less resilient, and more vulnerable to disease.
Image by Dr. David A. Sinclair, AO, PhD
From broken parts to broken instructions
The modern story of the ITOA began not with humans, monkeys, or even mice, but with microscopic yeast. Yeast are single-celled fungi – the same kind used to make bread and beer – and they’ve long been a workhorse of genetics because they grow fast and are easy to genetically manipulate.
In the mid-1990s, researchers at Massachusetts Institute of Technology, led by Dr. Leonard Guarente, sought to figure out why yeast cells age and how to slow this process down. The lab noticed a mutant yeast strain that survived long periods of starvation and, even more surprising, lived about 30% longer than normal cells. The mutant, called SIR4-24, pointed the lab toward a group of genes named “silent information regulators” (SIRs). These genes were already known to keep certain stretches of the yeast genome transcriptionally “quiet,” meaning those regions of DNA were tightly packed and their genes turned off. Now they seemed to be doing something else: determining how long the cells could live.
Zoom in on one of these silent information regulators, Sir2. In young cells, Sir2 keeps certain “sex” genes quiet so that the cell stays clearly one sex or the other, rather than a mix of both. It also sits with Sir4 and other partners on tightly packed DNA, including a fragile stretch that tends to break and form extra DNA loops. These loops, called extrachromosomal ribosomal DNA circles (ERCs), build up in old yeast cells and contribute to their death.4 Sir2 is the catalytic workhorse of this silencing complex: by chemically modifying the proteins that spool DNA, it keeps specific regions of the genome quiet and preserves the cell’s identity. Cells with extra Sir2 form fewer circles and live longer in standard lab conditions while cells without Sir2 accumulate ERCs sooner and die earlier.5
Between 1994 and 1999, work from the Guarente lab uncovered a pattern: when DNA damage occurs, Sir2 and its partners are pulled away from their usual posts on silent chromatin and sent to breaks in the genome to help with repair. DNA in cells breaks all the time, even under normal conditions. In young cells, Sir2 typically returns to where it started once the damage is fixed, but the system is not perfect. Each time Sir2 moves away, regions of DNA it was keeping tightly wrapped can loosen, genes that should stay silent can turn on, and over many cycles of damage and repair Sir2 does not always find its way back. The DNA letters themselves remain mostly correct, what changes is how the cell reads them. Old yeast cells, with misplaced Sir2 and an abundance of ERCs, ultimately lose their former identity. Their function changes, they lose the ability to mate, and eventually die under the burden of extra DNA circles that soak up transcriptional machinery.
The Sir2 story drew attention because it is part of a conserved family of genes, the sirtuins, with close counterparts in humans. That meant these yeast experiments might be pointing to a deeper, more universal rule about how cells age.
The conceptual leap came when these pieces were put together. DNA breaks actively pull sirtuins off their normal jobs, eroding the epigenetic landscape a little each time. Over many repair cycles, that relocalization changes gene expression, chips away at cellular identity, and eventually drives aging. In this light, aging looks less like the simple buildup of unrepaired lesions and more like epigenetic side effects of the cell’s own repair processes. The DNA sequence can remain largely intact, yet the instructions layered on top of it drift. That shift, from broken parts to broken instructions, is what eventually crystallized into the ITOA.
From yeast to mammals: when repair erodes identity
Mammalian genomes and biology are vastly more complex than yeast, though similarities exist. As in yeast, mammalian chromatin regulators such as sirtuins are recruited to DNA breaks. They leave their usual posts at promoters and silent regions to help with repair and, in the process, change gene expression.6
To test whether such shifts in epigenetic information could cause aging, rather than simply accompany it, researchers at Harvard Medical School created the ICE model (“ICE” for “Inducible Changes to the Epigenome”).7 In these mice, targeted, easy-to-repair DNA breaks were introduced throughout the genome without increasing the overall number of mutations beyond what normal mice experience. The result was striking. Animals developed accelerated epigenetic age, disrupted chromatin architecture, and functional signs of aging – physical frailty, organ decline – much earlier than expected. Their underlying DNA sequence remained intact, but their epigenetic patterns resembled those of much older animals. In other words, the yeast story holds in a mammal: repeated activation of the DNA repair machinery gradually rewires the epigenetic landscape, and that loss of epigenetic information is enough to drive an aging-like state.
On the right is a mouse which has undergone inducible changes to it's epigenome (an "ICE" mouse). On the left is a mouse born on the same day. Image adapted from Yang et al., Cell 2023
Epigenetic “noise” and Waddington’s landscape
One way to visualize all of this is through a picture first drawn in the 1940s by the embryologist Dr. Conrad Waddington, a picture we now refer to as “Waddington’s landscape”. To represent a cell as it goes through epigenetic changes during embryonic development, he imagined a ball rolling down a sloped, hilly terrain. The landscape has peaks and valleys; each valley represents a different cell identity (e.g., skin, muscle, neuron) and the hills between them act as barriers. At the top of the hill is the ball – representing an early embryonic cell with no fixed identity. As the ball rolls downhill during development, it settles into one valley, maturing into a particular kind of cell and staying there.
In youth, the hills are tall and the valleys are deep. The boundaries between them are sharp enough that cells “know who they are.” A skin cell doesn’t suddenly act like a neuron because there is a high barrier between the skin valley and the neuron valley. This landscape is shaped by the epigenome, with its patterns of DNA methylation, histone modifications, and chromatin loops. These marks carve the valleys and build the hills, guiding cells into stable identities during embryonic development and keeping them there early in life.
With age, combined with toxins and radiation that causes DNA to break, the landscape changes. Early in adulthood, the hills start to erode, the valleys become shallower, and the borders between them soften. Epigenetic marks drift and accumulate noise, making the overall shape of the landscape less distinct. In midlife, there appears to be a more rapid shift in age and cells metaphorically begin to slide sideways into neighboring valleys. The result is a partial loss of cell identity: liver cells express genes they shouldn’t, immune cells respond in the wrong way, and neurons don’t quite behave like neurons. The aches, the memory loss, and the diseases begin. It’s slow at first, but late in life, it’s a catastrophic decline.
Waddington’s landscape also helps frame how aging and age reversal might work. Shortly after fertilization, there is evidence that the epigenetic clock of embryos actually decreases until it reaches a “ground zero” of biological age.8 This finding suggests that a rejuvenation event occurs naturally during the lifespan, thus providing a framework to consider age reversal. A similar process seems to help lizards, salamanders, and fish regrow tails and limbs, suggesting perhaps this technology will go beyond aging and into organ and tissue regeneration.
Even if we don’t know exactly how epigenetic restoration works yet, we know that biology can rebuild a youthful epigenetic landscape. The ITOA predicts that similar principles could be understood and harnessed later in life.
Shannon, backups, and the “youthful code”
If aging really is an information problem, it helps to ask how information survives in noisy systems at all. Mathematician Dr. Claude Shannon, a pioneer in Information Theory that led to the digital world we live in, tackled a version of this in the 1940s: how do you send a message through a channel that introduces random errors and still have it arrive intact? This was important for sending radio signals efficiently and for them to be understood by the receiver. His answer was that you don’t eliminate errors, you correct them with a back-up. By building redundancy into the message and comparing what comes in against a clean reference, the receiver can spot where bits have been altered and reconstruct the original signal. Shannon formalized the idea that you can use a reference, sometimes framed as an “observer”, to detect and correct errors.
The ITOA borrows this logic. If features of old cells can be restored to a youthful state, despite decades of noise and damage, then perhaps somewhere there may be a reference that defines “youth”. Implicit in the ITOA is a suggestion that cells retain a latent backup of their youthful epigenetic state, a distributed record of which genes should be on or off, and how the epigenome should be arranged. That backup is not a single molecule but a web of molecular and structural cues specifying a youthful configuration of gene expression and chromatin architecture. Resetting it is complicated, involving thousands of genes and three-dimensional rearrangements of DNA, and we do not yet know where the reference copy resides, if it exists at all. Still, the fact that embryos reset their age and epigenetic restoration can push adult cells toward youth suggests that biology already knows how to recover that information. The challenge is to work out how embryos, lizards, and salamanders carry out this reset, and to harness those same pathways in a controlled way.
Reprogramming, with the volume turned down
If cells really do carry a backup of their youthful state, the obvious next question is whether we can access it in a safe way. The first clues that cells can be reprogrammed came from Dr. Shinya Yamanaka, who showed in 2006 that turning on four genes, Oct4, Sox2, Klf4, and c-Myc (OSKM), now called the Yamanaka factors, could “reprogram” mature cells back to a pluripotent, stem cell called an iPSC.9
That 2006 discovery ultimately won Yamanaka the Nobel Prize, but it also came with a problem: fully reprogrammed cells completely forget their identities. In animals, turning on OSKM can drive lethal tumors.10 In early mouse experiments, animals died within days of continuous OSKM expression. It wasn’t clear whether this was age reversal or simply a slow-motion catastrophe. And if reprogramming meant erasing cell identity, causing cancer or leading to death, how could it ever be used on patients?
For years, most scientists assumed that all four Yamanaka factors were required for reprogramming. Around 2009, several groups began exploring whether a subset of factors might induce a gentler form of reprogramming – enough to rejuvenate cells without erasing their identity, more of a restoration of the epigenetic landscape. After multiple combinations of embryonic genes and culture conditions failed to strike that balance, a surprising result emerged: dropping c-Myc and using just Oct4, Sox2, and Klf4 (OSK) could decrease epigenetic age without making them stem cells or causing tumors.11 In mice, brief OSK expression restored vision after optic nerve damage and reversed signs of aging in tissues such as muscle and kidney, including in ICE mice whose epigenome had been artificially aged by repeated DNA breaks.7 In the last 10 years, OSK has since been applied in several contexts, showing benefit in reversing features of aging, injury and disease across experimental models.
When epigenetic restoration leaves the lab
The harder challenge is translating a concept like this from the lab bench to the clinic, something only a tiny fraction of biological ideas ever achieve. To move beyond mice, researchers at Harvard Medical School, Massachusetts Eye and Ear Institute, and Life Biosciences used gene therapy to deliver OSK to retinal ganglion cells in non-human primates with optic-nerve injury and glaucoma-like damage. Treated animals showed measurable improvements in visual function, and their optic nerves exhibited DNA-methylation patterns that shifted toward those of younger monkeys. Importantly, in these early studies, the cells retained their identity as retinal neurons and no obvious safety signals emerged over the observation period, although longer and larger studies will be needed to rule out late effects and rare events.
Those results set the stage for ER-100, which has now received FDA approval for Phase I trials in people with glaucoma and non-arteritic anterior ischemic optic neuropathy (NAION), a condition caused by an acute reduction in blood flow to the optic nerve. These initial trials will test safety and dosing, and it remains to be seen how much functional recovery they will deliver. Even so, they represent something new: the first clinical trial of epigenetic restoration as a way to intervene in human aging and disease.
From theory to expectations
One reason the ITOA has traction in the field is that it offers a unifying way to think about a set of otherwise disconnected observations, without claiming to be the final word on aging.
For example, human aging follows a broadly similar script – graying hair, arterial stiffening, wrinkled skin, slower wound healing – despite wide genetic variation between individuals. In a purely damage-based view, this is puzzling: if damage accumulates randomly, why do so many people converge on the same symptoms late in life? With an ITOA framework, it’s because what we share is not identical DNA sequences, but highly conserved epigenetic and repair systems. The same broad families of chromatin regulators, DNA-methylation enzymes, and damage-response pathways operate in all of us. If those shared systems lose information in stereotyped ways, they will tend to fail along similar trajectories, giving rise to familiar downstream problems: chronic inflammation, cellular senescence, mitochondrial dysfunction, stem cell exhaustion, and other “aging hallmarks” that show up across individuals and even across species.
The ITOA also connects naturally to the rise of epigenetic clocks, which are able to track chronological age reasonably well and can forecast risk of disease, disability and death across studies. If aging is the progressive loss of structured epigenetic information, then epigenetic patterns, such as DNA methylation, should provide a sensitive measure of how far an individual or organ has drifted from its youthful state. Newer “next-generation” clocks push this further by adding clinical data and other molecular readouts on top of methylation, and they seem to do a better job predicting healthspan and disease risk.12 That is what an information-based view would predict: the more aspects of the system you measure, the more precisely you can place someone along their aging trajectory.
Perhaps most importantly, the ITOA provides a natural language for thinking about rejuvenation itself. Nuclear transfer experiments, when scientists place the nucleus of an old cell into an egg, can support development of a healthy young animal, showing that a “chronologically old” genome can, in the right context, give rise to young tissues. Epigenetic restoration with OSK, at least in current rodent and primate studies, can restore function in aged or injured tissues without obliterating cell identity. Together, these findings imply that the genome is not irreversibly aged in the way a rusted car frame is. Rather, the instructions for using that genome have been modified, and some aspects of those modifications are reversible.
Even so, the ITOA is not a finished “general theory.” The aging field still lacks a framework that can, on its own, explain the wide variety of aging patterns seen across species, the effects of diverse longevity mutations, and why some interventions compress morbidity while others simply stretch late-life decline. Can epigenetic restoration address all of the famous Hallmarks of Aging, or only a subset? How many times is reset possible? The ITOA is one attempt to knit those observations together, but it will stand or fall on how well its predictions hold up in experiments and, eventually, in people.
Therapies for an information-based view of aging
The ITOA is woven into the logic behind ER-100’s design: if aging reflects corrupted epigenetic instructions, then carefully nudging those instructions back toward youth might help damaged tissue recover. The Phase I trials in glaucoma and NAION are small, and focused mainly on safety but, whether these first trials reverse blindness or not, they mark a turning point in medicine. Rather than simply slowing further damage, the goal is to see whether epigenetic restoration can restore function in tissue that is already old or injured. Regardless of outcome, these studies will tell us something important about how far the epigenetic landscape in human tissue can be pushed and the risk that entails.
If ER-100 shows even modest benefit, the obvious next question is where else this technology might apply. Therapeutic epigenetic restoration could in principle be adapted to the liver, muscle, or nervous system, to target diseases that are currently considered irreversible. In an ITOA framework, the goal is not just to patch one pathway at a time, but to restore cell identity and function by resetting the information that defines them. The hope is that, once that information is reset, cells can prevent and resolve disease on their own.
How we deliver that reset will inevitably diversify. Viral vectors are the first wave, but several groups are exploring alternative delivery routes, including lipid nanoparticles and small-molecule cocktails. At the same time, a growing number of companies – Life Biosciences, Retro Biosciences, Altos Labs, Rejuvenate Bio, NewLimit, and others – are racing to turn these concepts into medicines, backed by several billion dollars in investment and a vast R&D ecosystem. The approaches differ, but they share a common bet: if aging is fundamentally an information problem, then therapies that restore the right information could, in principle, treat many age-related diseases at once, rather than tackling them one pathway and one organ at a time.
Standing at the threshold
We return to the exam room, where the patient is still speaking with his doctor. The chart on the wall next to him remains blurred at the edges. Whatever happens next, something remarkable has already occurred: a person with glaucoma is being offered a clinical therapy which attempts to target cellular age.
Reaching this moment has taken far more than one person, one lab, or one theory. It grew from the slow realization that aging might be, at its core, a problem of lost biological information. Transforming that idea into testable models, experimental tools and ultimately into a clinical-grade therapy has required decades of work, thousands of experiments, and the sustained effort of many scientists, funders, and institutions.
Whether ER-100 ultimately benefits this patient is what the early trials aim to determine. What is already evident is that the field has crossed a threshold. We are no longer only documenting the erosion of cellular information. We are beginning to explore whether it can be restored. For the patient in the exam room, and for everyone who will grow old, that shift marks the start of a new phase in aging research, where the information in our cells is not only something to map, but something we may one day recover.
References
1. de Magalhaes, J. P. An overview of contemporary theories of ageing. Nat Cell Biol 27, 1074-1082, doi:10.1038/s41556-025-01698-7 (2025).
2. Weinert, B. T. & Timiras, P. S. Invited review: Theories of aging. J Appl Physiol (1985) 95, 1706-1716, doi:10.1152/japplphysiol.00288.2003 (2003).
3. Laughrea, M. On the error theories of aging. A review of the experimental data. Exp Gerontol 17, 305-317, doi:10.1016/0531-5565(82)90020-1 (1982).
4. Sinclair, D. A. & Guarente, L. Extrachromosomal rDNA circles--a cause of aging in yeast. Cell 91, 1033-1042, doi:10.1016/s0092-8674(00)80493-6 (1997).
5. Kaeberlein, M., McVey, M. & Guarente, L. The SIR2/3/4 complex and SIR2 alone promote longevity in Saccharomyces cerevisiae by two different mechanisms. Genes Dev 13, 2570-2580, doi:10.1101/gad.13.19.2570 (1999).
6. Oberdoerffer, P. et al. SIRT1 redistribution on chromatin promotes genomic stability but alters gene expression during aging. Cell 135, 907-918, doi:10.1016/j.cell.2008.10.025 (2008).
7. Yang, J. H. et al. Loss of epigenetic information as a cause of mammalian aging. Cell 187, 1312-1313, doi:10.1016/j.cell.2024.01.049 (2024).
8. Kerepesi, C., Zhang, B., Lee, S. G., Trapp, A. & Gladyshev, V. N. Epigenetic clocks reveal a rejuvenation event during embryogenesis followed by aging. Sci Adv 7, doi:10.1126/sciadv.abg6082 (2021).
9. Takahashi, K. & Yamanaka, S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 663-676, doi:10.1016/j.cell.2006.07.024 (2006).
10. Abad, M. et al. Reprogramming in vivo produces teratomas and iPS cells with totipotency features. Nature 502, 340-345, doi:10.1038/nature12586 (2013).
11. Lu, Y. et al. Reprogramming to recover youthful epigenetic information and restore vision. Nature 588, 124-129, doi:10.1038/s41586-020-2975-4 (2020).
12. Johnson, A. A. & Shokhirev, M. N. First-generation versus next-generation epigenetic aging clocks: Differences in performance and utility. Biogerontology 26, 121, doi:10.1007/s10522-025-10265-4 (2025).
Lifespan Magazine is proud to support medical research and the Lifespan Foundation
Last Updated: Aug 7, 2026 · 19 minute
