Does SRSF1 reverse ageing? In old human cells in a dish and in aged mice, boosting this one gene shifted cells toward a younger state and extended median lifespan by 14%. In people, it has not been tested at all. That gap, between a striking laboratory result and any human evidence, is the whole story of the October 2026 Nature Aging paper that put SRSF1 on the longevity map.

The quotable version: SRSF1 is a real, peer reviewed lead, not a product. A Harvard led team used a gene expression age predictor to screen for genes that could rejuvenate old cells without erasing what those cells are, and SRSF1 came out on top. It reduced markers of cellular senescence, protein stress and inflammation in aged human fibroblasts, and in aged mice it slowed frailty and lengthened life. No human trial exists, no SRSF1 therapy or supplement is available, and the delivery method used in mice targeted mainly the liver. Anyone selling you an SRSF1 product today is selling something the science does not support.

What SRSF1 actually is

SRSF1 stands for serine and arginine rich splicing factor 1. It is an RNA processing factor, and its best known job is alternative splicing: helping decide which protein coding sections of a messenger RNA molecule are kept when the final message is assembled. Because one gene can produce several different proteins depending on how its message is spliced, a splicing factor can quietly rewire a large part of a cell’s behaviour without changing the DNA itself.

That is why the Nature Aging authors, led by George Church at the Wyss Institute at Harvard University with co first authors Alexandru M. Plesa and Sascha Jung and collaborators at CIC bioGUNE in Spain, call SRSF1 a regulator of ageing rather than a simple anti ageing switch. The paper, titled “Transcriptomic reprogramming screen identifies SRSF1 as a regulator of aging”, was published in Nature Aging on 14 October 2026. Its central claim is not that one gene magically makes you young. It is that boosting SRSF1 shifts the gene expression programmes of old cells toward younger patterns, across several hallmarks of ageing at once.

How the team found it: an age predictor built for screening

Most ageing clocks measure chemical marks on DNA, known as DNA methylation. Those clocks are useful, but they are expensive to run at screening scale and hard to interpret when you want to test many candidate genes. The Harvard team took a different route. They built a machine learning predictor trained on gene expression, or transcriptomic, profiles from 464 human skin fibroblast samples from donors across a wide age range. Instead of one overall age number, the model tracks separate processes, rather like odometers for different parts of a car: protein quality control, translation, actin organisation, histone methylation and endoplasmic reticulum stress, among others.

The predictor was checked against known lifespan changing interventions, including the Yamanaka factors, in fibroblast lines from donors aged 55, 65 and 79. It reported the expected shifts in transcriptomic age. That gave the team a scorable readout: overexpress a candidate gene in an old fibroblast line, and ask whether the predictor now scores that line as younger. They screened a focused library of genes predicted to affect ageing, and SRSF1 scored highest.

What happened in old human cells

In aged human fibroblasts, SRSF1 overexpression reduced features corresponding to several hallmarks of ageing at the same time. Cellular senescence fell: fewer of the damaged cells that stop dividing but refuse to die, and that secrete inflammatory signals into surrounding tissue. Proteostatic stress eased: the cell’s protein balance and turnover, which deteriorates with age, moved toward a younger pattern. Inflammatory signalling, the chronic low grade inflammation often called inflammaging, also fell. DNA repair programmes were restored, and the rejuvenation registered on an epigenetic ageing clock as well as on the transcriptomic predictor.

The detail that matters most is what did not happen. The rejuvenated fibroblasts kept their identity. They retained their characteristic spindle shape, kept expressing genes for typical fibroblast functions, and did not switch on the pluripotency programmes that define stem cells. That is the point the team keeps returning to, in the words of co first author Plesa: in reprogramming, there is a fundamental tradeoff between rejuvenation and cell identity, and achieving one without losing the other requires walking a fine line.

Why keeping cell identity matters: the Yamanaka problem

The comparison point is Shinya Yamanaka’s 2006 discovery that four genes, the Yamanaka factors, can reprogram mature cells back to a pluripotent ground state. Expressed only in short bursts, those factors can partially reprogram old cells to a more youthful state while keeping their specialised functions, which proved that ageing is not necessarily a one way street. But full reprogramming erases identity, and overexpression of the Yamanaka factors raises concern about cancer formation. That is why researchers have hunted for factors that rejuvenate without pushing cells all the way back to stemness.

SRSF1 is attractive precisely because it works through machinery every cell already uses, rather than overriding identity from outside. It does not reprogram a skin cell into something else. It nudges the splicing and expression programmes of that skin cell toward the patterns seen in younger skin cells. That is a narrower, more controllable claim than full reprogramming, and it is also why the result is more credible as a research lead, even though it is much further from a therapy.

Does SRSF1 reverse ageing in people? The mouse answer and the human gap

Does SRSF1 reverse ageing in mice? On the reported measures, yes, and by a margin worth noting. The team packaged the SRSF1 gene in a liver tropic adeno associated virus, or AAV, and treated aged mice. The therapy accumulated mainly in the liver, yet the benefits were systemic. SRSF1 treated animals grew frail more slowly and lived longer, with median lifespan extended by 14% compared with controls.

The frailty numbers are more informative than the lifespan headline. In the paper, SRSF1 induction slowed the rate of frailty accumulation by about 44%: control mice accumulated deficits at 0.086 per week, while treated mice accumulated them at 0.048 per week. That flattened trajectory delayed the onset of sustained frailty by about 24%, or roughly 123 days. Mortality risk was also significantly lower in treated mice, with a hazard ratio of 0.07 reported in the paper. Liver DNA methylation age acceleration fell by 0.5 years on the intervention optimised clock, although that effect was not significant on a pan tissue clock and was not detected by a chronological liver clock.

Read those numbers with the same care the authors invite. Improvements in individual frailty components, such as breathing rate and skin condition, did not remain significant after correction for multiple comparisons. The systemic pattern, slower decline across many systems rather than a dramatic rescue of one organ, is the finding. And secondary reporting on the study describes the mouse work as a small cohort. A 14% median lifespan extension in a small, liver targeted mouse study is a reason to run bigger studies. It is not a reason to extrapolate to humans.

What this study does not prove

First, there is zero human evidence. No human trial of SRSF1 gene therapy has been reported, and no dose, delivery route or safety profile in people exists. Mouse lifespan does not translate to human lifespan in any simple way, and liver targeted AAV delivery in mice is a long way from a safe, general human therapy.

Second, the mechanism is only partly understood. The authors report a strong trend of alternatively spliced messenger RNAs, but, in Plesa’s own words, they are only just starting to understand how those splicing changes bring about the multifaceted rejuvenation programme in tissues. A lead whose mechanism is incomplete is normal at this stage of research. It is also a reason not to assume the effect will hold in other tissues, other ages or other species.

Third, delivery and dosing are unsolved problems, and the authors say so directly. Solving delivery and dosing could turn this geroprotective gene into a rejuvenation therapy, they argue, but that sentence contains the whole challenge. Getting a gene therapy to the right tissues, at the right level, for the right duration, without unwanted effects, is where most promising mouse findings stall. Nothing in this paper removes that obstacle.

Fourth, do not confuse a transcriptomic age score with being younger in any clinical sense. The predictor measures gene expression patterns. Those patterns moved toward younger profiles, and in mice that movement tracked with slower frailty and longer life, which strengthens the case. But a younger score on a laboratory predictor is not the same as a younger body, and it is not a validated surrogate for human health outcomes.

The honest biohacker take: what, if anything, should you do differently?

Nothing you can buy, take or do today follows from this paper, and that is the honest answer. There is no SRSF1 supplement, no clinic offering SRSF1 gene therapy, and no at home test that measures your SRSF1 activity in a clinically meaningful way. Any product using the SRSF1 name is borrowing the credibility of a paper that tested none of it.

What does follow is a better filter for the next longevity headline. Ask four questions before you care about a new ageing gene. Was the result shown in human cells, in animals, or in people? Did the intervention preserve cell identity, or did it push cells toward stemness? Did it move a functional outcome, like frailty or strength, or only a molecular score? And who funded it, and how large was the study? SRSF1 passes the first three filters better than most October headlines: human cells plus a functional mouse outcome, identity preserved, and funding from the Wyss Institute, the Aging and Longevity Related Research Fund, the Astera Fund and the Spanish Ministry of Science and Innovation. It fails the fourth in the sense that matters most to you: it has not been tested in people at all.

The sensible protocol, if you can call it that, is to watch the pipeline rather than chase the molecule. The next informative steps would be replication in larger mouse cohorts, studies in other tissues and delivery routes, and a clear safety picture over a full lifespan. Until those exist, the interventions with actual human evidence for healthy ageing remain the unglamorous ones: regular exercise, sleep, not smoking, and managing blood pressure and metabolic health. SRSF1 may one day join that list through a very different route. Today it belongs on a watchlist, not in a shopping basket.

Frequently asked questions

Does SRSF1 reverse ageing?

Does SRSF1 reverse ageing in old human cells and aged mice? Yes, on the measures reported: younger gene expression patterns, less senescence and inflammation, slower frailty and 14% longer median lifespan in mice. In people, the question is untested, so the honest answer is that we do not know.

What is SRSF1?

SRSF1 is a serine and arginine rich splicing factor, an RNA processing protein that helps decide which sections of a messenger RNA are kept. By changing splicing patterns, it can shift many gene expression programmes at once without altering the underlying DNA sequence.

Is SRSF1 available as a supplement or treatment?

No. SRSF1 was delivered in mice as a gene in a liver tropic virus, not as a pill or supplement. No human therapy exists, and any product marketed under the SRSF1 name has not been tested in the study.

How is SRSF1 different from the Yamanaka factors?

The Yamanaka factors can reprogram mature cells back to a pluripotent, stem cell like state, which carries cancer risk when overexpressed. SRSF1 rejuvenated old fibroblasts while they kept their identity, shape and specialised gene expression, and did not switch on pluripotency programmes.

How much longer did the treated mice live?

Median lifespan was extended by 14% compared with controls. Frailty accumulated about 44% more slowly, and the onset of sustained frailty was delayed by roughly 123 days. These are mouse results from a liver targeted study, not human forecasts.

Should biohackers try to raise their SRSF1 levels?

There is no validated, safe way for a person to do this, and no evidence that any diet, supplement or lifestyle intervention raises SRSF1 activity in a therapeutically meaningful way. The study supports further research, not self experimentation.

Sources

Plesa, A. M., Jung, S., et al. “Transcriptomic reprogramming screen identifies SRSF1 as a regulator of aging.” Nature Aging, 14 October 2026. DOI: 10.1038/s43587-026-01234-8. Nature Aging paper. Wyss Institute at Harvard University. “Reversing age by reprogramming gene expression networks.” Wyss Institute news. News-Medical. “Novel gene rejuvenation therapy preserves cell identity while reversing age.” 14 October 2026. News-Medical report.

Related reading

Does Urolithin A Actually Do Anything for Muscle? Do GLP-1 Drugs Really Reverse Biological Ageing? Can You Train Processing Speed, the Brain Skill Tied to Longer Life?

Disclaimer: this article explains published research and is not medical advice. Gene therapy and longevity interventions should not be attempted outside regulated clinical research.