Harvard Scientists Just Mapped the Universal Molecular Clock of Aging

A landmark Nature study from Mass General Brigham identifies gene expression signatures so fundamental to biological aging that they appear nearly identically across mice, rats, macaques, and humans, opening a new era for longevity medicine.

Presented By Our Partners

For the first time in the history of aging science, researchers have identified a set of molecular signatures so conserved across evolution that they appear in nearly identical form in species separated by 90 million years of divergence. The finding, published June 4, 2026 in Nature by a team led by investigators at Mass General Brigham, introduces a new class of biological tools called transcriptomic clocks that can measure not only how old an organism is, but how quickly it is approaching death, and which specific biological processes are driving that decline.

The implications reach far beyond academic biology. If aging has universal molecular signatures, then the science of slowing or reversing it becomes dramatically more tractable. Researchers can now measure interventions, from calorie restriction to new drugs, against a precise biological yardstick rather than waiting decades for mortality outcomes. It is the kind of foundational advance that longevity science has needed for a generation.

The Question That Has Defined Aging Science

Why do we age? The question sounds simple. The answer has occupied some of the sharpest minds in biology for more than a century. For most of that time, aging research was fragmented: different tissues aged differently, different organisms aged at different rates, and what happened in a mouse often did not translate to a human.

The fundamental problem was always measurement. Without a reliable way to quantify biological age independently of calendar age, it was impossible to know whether an intervention was genuinely slowing aging or merely producing a superficial health effect. DNA methylation clocks, pioneered by Steve Horvath in 2013 and refined extensively since, represented a major breakthrough in this direction. But methylation clocks measure one molecular layer. Gene expression, the actual activation and suppression of genes in living tissue, tells a richer and more dynamic story about what is actually happening inside cells as they age.

“We found that most cell types share these conserved molecular changes with age, despite having very different origins and functions, from immune cells and stem cells to liver cells and muscle cells,” said lead author Alexander Tyshkovskiy, PhD, an investigator in the Division of Genetics in Mass General Brigham’s Department of Medicine. “The same biomarkers were also predictive of time to death in humans and responsive to chronic diseases and lifespan-modulating interventions, such as calorie restriction, in mice.”

The paper, titled “Universal transcriptomic hallmarks of mammalian ageing and mortality,” was published in Nature (Vol. 654, pages 173-188, 2026), with DOI: 10.1038/s41586-026-10542-3. The research was funded by the National Institute on Aging, Hevolution, the James Fickel and Michael Antonov Foundations, and several international science agencies.

Inside the Largest Gene Expression Dataset in Aging Research

To build their transcriptomic clocks, Tyshkovskiy, senior author Vadim N. Gladyshev, PhD, and their collaborators assembled something that had never existed before: a comprehensive cross-species, cross-tissue gene expression atlas of biological aging.

The dataset includes more than 11,000 transcriptomes, detailed snapshots of which genes are active and at what levels, drawn from more than 25 distinct tissues across four mammalian species: mouse, rat, macaque, and human. Collaborators at Tohoku University in Japan contributed critical primate and rodent data to the analysis. The breadth of tissues sampled was intentional. By examining gene expression simultaneously in immune cells, stem cells, liver cells, muscle cells, brain tissue, and more, the team could determine which aging signatures were universal versus which were confined to specific tissue types.

Crucially, the researchers analyzed gene expression not only across age but across interventions known to extend or shorten lifespan. Calorie restriction, one of the most robustly validated lifespan-extending interventions in model organisms, appeared clearly in the molecular data. So did genetic manipulations that accelerate aging. Cellular stress induced by prolonged laboratory culture or radiation exposure produced molecular signatures strikingly similar to those seen in naturally aged tissues, revealing a direct mechanistic connection between cellular damage and organismal aging.

This design allowed the team to do something previous aging studies could not: separate gene expression changes that are merely correlated with aging from those that are causally linked to biological deterioration and elevated mortality risk. That distinction matters enormously for drug development and clinical translation.

Three Clocks, Three Questions

The research team did not build one clock. They built three, each answering a distinct question about biological aging, and each providing a different kind of clinically useful information.

The first clock estimates chronological age, how long an organism has been alive. This is conceptually similar to existing epigenetic clocks but grounded in transcriptomic rather than methylation data. Across species and tissues, the chronological transcriptomic clock performed with high accuracy, confirming that gene expression carries reliable age information independent of the epigenome.

The second clock, and arguably the more clinically significant of the three, measures expected mortality risk based on accumulated biological damage rather than elapsed time. Two individuals of the same calendar age can have dramatically different mortality risk based on how much cellular and tissue-level deterioration they have accumulated. The mortality clock captures this divergence precisely, offering a window into health trajectory that a birth date alone cannot provide.

Featured Partner

Invest in the Infrastructure Behind Modern Medicine

As healthcare expands beyond hospital walls, the buildings and campuses supporting that shift are generating compelling returns for investors who move early. The Healthcare Real Estate Fund offers qualified investors direct access to a curated portfolio of medical office, outpatient, and specialty care facilities.

Learn More →

The third clock expresses biological age as a percentage of total expected lifespan. This lifespan-relative measure allows meaningful comparisons across species that age at radically different rates. A two-year-old mouse is physiologically ancient; a two-year-old macaque is an infant. Expressing biological age as a fraction of expected total lifespan provides a unified biological currency for cross-species research and for evaluating interventions that may compress or extend the aging trajectory at different life stages.

Together, the three clocks provide a comprehensive molecular portrait of where an organism stands in its aging process, how much biological damage has accumulated, and how that damage compares to what would be expected for its species and lifespan. No prior research tool has offered all three dimensions simultaneously.

The Genes at the Center: GPNMB, CDKN1A, and LGALS3

Among the study’s most actionable contributions is the identification of specific genes whose expression changes with striking consistency across all four mammalian species tested. Three genes emerged as particularly robust universal hallmarks of aging and elevated mortality risk: GPNMB, CDKN1A, and LGALS3. A fourth gene, CST7, also appeared as a consistent cross-species aging marker.

GPNMB, or glycoprotein non-metastatic melanoma protein B, is associated with lysosomal stress and macrophage activation. Its consistent upregulation with age across species suggests that cellular waste-processing systems are under progressive stress as organisms age, a finding that aligns closely with the broader literature on lysosomal dysfunction in aging and neurodegenerative disease. When lysosomes fail to efficiently clear damaged proteins and organelles, cellular aging accelerates.

CDKN1A encodes p21, one of the most studied cell cycle inhibitors in molecular biology. It is a central regulator of cellular senescence, the state in which cells permanently exit the cell cycle but remain metabolically active, secreting inflammatory signals that damage surrounding tissues. The universal upregulation of CDKN1A across aging mammals provides strong molecular evidence that cellular senescence is a conserved driver of biological aging rather than a species-specific phenomenon. This lends significant scientific weight to the growing field of senolytic therapies, which aim to selectively clear senescent cells.

LGALS3, encoding galectin-3, is a well-characterized mediator of inflammation and fibrosis. Its consistent increase with age across tissues and species points to chronic, low-grade inflammation, sometimes called “inflammaging,” as a fundamental conserved feature of mammalian aging. Elevated galectin-3 is already used in clinical settings as a biomarker of cardiac fibrosis and heart failure risk; its emergence here as a universal aging marker suggests its clinical utility may extend further.

CST7, encoding cystatin F, rounds out the quartet. It is involved in immune activation and lysosomal function, and its upregulation with age reinforces the picture of aging as a state of progressive immune dysregulation and cellular cleanup failure.

The convergence of these four markers around the same three biological themes, lysosomal stress, cellular senescence, and chronic inflammation, is not coincidental. These processes are mechanistically interconnected: senescent cells drive inflammaging through their secretory phenotype, and lysosomal dysfunction accelerates both senescence accumulation and inflammatory signaling. The transcriptomic data now provides cross-species molecular evidence that this triad is a conserved hallmark of mammalian aging.

Aging Is Not One Process: The Modular Approach

One of the most clinically significant aspects of the new research is its modular architecture. Rather than treating aging as a single undifferentiated process captured in one summary score, the team separated gene expression changes into functional modules representing distinct biological pathways.

The modules include inflammation signaling, mitochondrial energy production, extracellular matrix organization, immune function, and cellular stress response, among others. For each module, the team developed a separate transcriptomic clock. The result is a disaggregated, high-resolution view of biological aging that reveals not just how old someone’s biology is, but which specific systems are failing and at what rate.

This modular approach has profound implications for personalized medicine. Different chronic diseases appear to accelerate different modules. Cardiovascular disease accelerates the extracellular matrix and inflammation modules more than the mitochondrial energy clock. Metabolic disease shows a distinct signature in energy production and inflammatory signaling. Neurodegenerative disease maps onto different molecular pathways still. Matching a patient’s modular aging signature to a targeted intervention could represent a significant advance over the current one-size-fits-all approach to preventive medicine.

“Future therapies could target both specific aging-related processes, like inflammation or metabolism, and aging as a whole,” said senior author Vadim N. Gladyshev, PhD, a geneticist in the Division of Genetics at Mass General Brigham. “These aging clocks represent a potential new way to measure aging in greater detail and could help predict disease and mortality risk, characterize treatment effects, and personalize care based on biological age.”

The modular framework also helps explain why different interventions work through different mechanisms. Calorie restriction, metformin, rapamycin, and exercise each appear to affect distinct subsets of the aging modules. This provides a rational molecular basis for combination approaches that target multiple aging pathways simultaneously, one of the foundational concepts driving the next generation of longevity medicine.

Calorie Restriction, Visible in the Clock

Among the study’s most striking findings is the visibility of calorie restriction in the transcriptomic data. Calorie restriction, reducing food intake by roughly 20 to 40 percent without malnutrition, is the most robustly validated intervention for extending healthy lifespan in model organisms. It extends median and maximum lifespan in every species tested, from nematodes to primates, and consistently reduces incidence of the major age-related diseases.

The transcriptomic clocks built from the Mass General Brigham dataset are sensitive enough to detect calorie restriction’s biological signature. Animals on calorie-restricted diets showed measurably younger transcriptomic profiles than age-matched controls fed normally, and the difference appeared across multiple aging modules, most prominently in inflammation and metabolic signatures. This is the first time a transcriptomic clock has been shown to capture calorie restriction’s biological age-reducing effect at a multi-tissue, multi-species scale.

The study also examined genetic mutations known to extend lifespan in mice, including Ames dwarf mutations affecting growth hormone and IGF-1 signaling. These genetic interventions produced transcriptomic signatures consistent with reduced biological age, providing molecular-level validation for their observed longevity effects. Conversely, conditions that accelerate biological aging, including radiation exposure and cellular senescence induced by prolonged laboratory culture, produced transcriptomic signatures that closely resembled aged tissue even in cells that were chronologically young. The data show that cellular damage drives biological aging signatures independently of elapsed time.

Validated Against 75 Years of Human Health Data

A critical step in any biological clock’s development is validation against real-world human outcome data rather than laboratory models alone. The team tested their transcriptomic mortality clocks against the Framingham Heart Study, one of the most comprehensive and long-running cardiovascular cohort studies in medical history, tracking participants across decades for mortality, disease incidence, and health trajectories.

The transcriptomic clocks successfully predicted mortality risk in the Framingham cohort. Their performance was comparable to, and in several analyses superior to, existing DNA methylation-based aging clocks. This validation against decades of human health data provides meaningful evidence that the transcriptomic approach is capturing biologically real aging signals rather than computational artifacts of the large dataset.

The research team has made their tools publicly available to the scientific community for non-commercial use through an interactive web platform and an R package, enabling other researchers to apply the clocks to new datasets and accelerate validation, refinement, and clinical translation.

What This Means For You

The discovery of universal transcriptomic hallmarks of aging is, first and foremost, a landmark scientific result. But its implications extend directly to anyone invested in their own healthspan and biological longevity.

The three biological processes that dominate the universal aging signature, chronic inflammation, cellular senescence, and lysosomal dysfunction, are not inevitable fates. They are processes that foundational health practices measurably influence. Calorie restriction appears in the transcriptomic clock as a genuine biological age reducer. Exercise has been shown in multiple peer-reviewed studies to reduce markers of inflammaging and cellular senescence burden, effects that would predictably register in the modular transcriptomic clocks. Sleep quality profoundly affects inflammatory signaling. Fasting protocols activate autophagy, the cellular cleanup mechanism that governs lysosomal health.

The modular architecture of the clocks also suggests that different people may benefit from different prioritized interventions. Someone whose biological aging is dominated by the inflammation module may benefit most from anti-inflammatory lifestyle strategies: dietary quality, sleep optimization, stress management, and resistance training. Someone whose mitochondrial energy module is aging fastest may respond more strongly to aerobic exercise, certain nutritional strategies, and potentially emerging mitochondrial-targeted compounds.

We are not yet at the point where a clinician can order a transcriptomic clock panel and receive a personalized aging report. That day is meaningfully closer than it was before this study. What we can do now is act on the biological knowledge the clocks have revealed: the universal enemies of healthy aging are inflammation, senescence, and cellular stress, and the foundational practices of nutrition, movement, sleep, and stress management target all three.

The universal molecular clock of aging has been found. The levers that influence it have been known, practiced, and validated for years. The gap between that knowledge and daily practice remains the most important one to close.

Source: Tyshkovskiy A, et al. “Universal transcriptomic hallmarks of mammalian ageing and mortality.” Nature. 2026;654:173-188. DOI: 10.1038/s41586-026-10542-3. Research led by Alexander Tyshkovskiy, PhD, and Vadim N. Gladyshev, PhD, Division of Genetics, Mass General Brigham Department of Medicine.

Free Daily Briefing

The Latest Longevity Science.
Delivered Every Morning.

Join researchers, physicians, and health professionals getting daily breakthroughs in AI-driven medicine, epigenetics, and longevity research.

Support the research that powers this editorial

Subscribe Free

No spam. Unsubscribe anytime. We respect your inbox.