← Longevity & Supplement Guides
Active Is Not Trained: What Half of Muscle Ageing Turned Out to Be
Sixteen people in their late sixties had trained three times a week, an hour at a time, without a break, for more than a year. Fifteen others of the same age walked as many steps as the twenty-somethings in the study and logged the same time at higher intensities — by any wearable's reckoning, they were doing fine. Then a needle took a small piece of thigh muscle from each of them, and researchers read what the tissue was actually doing.
The results, published in Nature Aging in July, have been reported as exercise rolling back muscle ageing. That is the fun version. The quieter version is more useful, and it turns on a distinction almost nobody makes: being active and being trained are not the same condition.
What the study actually did
The team, working across Amsterdam UMC and Maastricht University, recruited forty-seven people into four groups: eleven young adults in their twenties, fifteen normally active older adults, sixteen trained older adults, and five older adults with early physical impairment. "Trained" had a specific meaning — at least three supervised hour-long sessions a week, sustained without interruption for over a year.
The design choice that makes the paper worth reading is the second group. Older people usually move less than young people, so any comparison of old and young muscle is contaminated from the start: you cannot tell ageing from disuse. So the researchers recruited older adults whose daily step counts and higher-intensity minutes matched the young adults'. On the outside, those two groups were living the same physical week.
Then they read the muscle's genes, lipids and metabolites before and after a single bout of submaximal cycling — the resting state of the tissue, and how it answered a demand.
Half the signature was simply absent
In the normally active older adults, more than a thousand genes were expressed at lower levels than in the young, and the list was dominated by cellular respiration — how muscle turns fuel into usable energy. This is the familiar picture of muscle ageing, and it appeared in people who were walking as much as their grandchildren.
In the trained older adults, much of that picture was missing. Roughly 56% of the age-related increases in gene expression and 57% of the decreases were not present, with estimates across the analyses ranging from about 45% to 62%. Their profiles for energy metabolism sat close to the young adults'. Youthful expression was maintained in NDUFS1, in COX5A, in ATP synthase subunits — across all five complexes of the respiratory chain.
Put plainly: about half of what we file under "muscle ageing" in this cohort looked less like ageing than like the absence of a training stimulus.
Evidence
Same weekly movement, different muscle
Answer to one exercise session, against the young adults'
Which parts of the ageing signature travelled with training — and which did not?
Close to young levels in the trained group: cellular respiration and energy metabolism, including NDUFS1, COX5A and ATP synthase subunits spanning all five respiratory complexes.
Age-shifted regardless of training: genes of synaptic transmission such as PCDH8 and UNC13C, and WNT signalling genes including DAAM2 and CTR9 — the wiring and remodelling side rather than the power side.
The paper is cross-sectional, so these are differences between groups, not changes measured inside one person over time.
Steps measure movement, not training
The daily step count has become the folk measure of whether a person is looking after themselves. It has earned some of that: the association between more steps and lower mortality is one of the more robust findings in the field, and worth understanding on its own terms. But it answers a narrower question than we ask of it. Both older groups moved. Only one of them had the muscle.
There is something faintly comic about the affluent world's arrangement here — a decade spent learning to close a ring, in the belief that the ring was the thing. It was never the thing. It was a proxy for the thing, and proxies drift.
What the trained group had was not more movement. It was a repeated, structured demand that the body had to build for. That is the argument behind Zone 2 time, resistance work after forty, and the finding that intensity, not weight change, tracks what happens to older muscle. Movement keeps you from decaying. Training asks a question the tissue has to answer.
What training did not touch
The honest half of the paper is the part that did not come along. Genes governing synaptic transmission and WNT signalling stayed age-shifted whether or not people trained — the nerve-to-muscle interface and the remodelling around it. Muscle does not only weaken with age; it also loses some of its wiring, and this cohort gives no indication that three sessions a week addressed that.
Which is the correct shape for a finding like this. Training is not a solvent that dissolves ageing. It appears to hold the energetic machinery of muscle near where it was, while other parts of the tissue keep their own schedule. Half is a remarkable share. It is also half. The rest of the story lives in what muscle does as an organ rather than as an engine.
The fit responded harder to one session
The acute part of the experiment produced the finding I find most interesting, and the one least likely to make headlines. Every group mounted an immune and stress transcriptional response to a single bout of cycling. The size of that response tracked physical function: the better a participant's fitness, the more their muscle behaved like a young person's when asked to work.
Read that as a redefinition. We tend to think of fitness as a capacity to perform — how far, how fast, how heavy. This suggests it is also a capacity to respond: whether a stimulus still lands, whether the tissue can still hear it. That is a more useful way to hold a number like VO₂max, or the speed of your heart rate's return after a hard effort.
What forty-seven people cannot tell you
Now the caveats, which are substantial. This is a cross-sectional study of forty-seven people, five of them in the impaired group; nobody was followed as they took up training. It reads bulk tissue, not individual cells. The cohort is Dutch and unusually active at baseline, so "normally active" here may be brisker than elsewhere. And the trained participants did a higher absolute workload in the test bout, so part of their larger response may simply be a larger stimulus.
Then the direction of the arrow. People who can train three hard hours a week for a year are already, in some respects, not the average sixty-eight-year-old. The study cannot separate "training kept this muscle young" from "muscle that ages slowly lets you keep training." Both readings fit. This is an association, described in tissue, and should be held as one.
The bottom line
Around half of the muscle-ageing signature in this cohort tracked with training status rather than with age, the energetic half most of all. Part of it did not budge. And the design cannot prove which way the causation runs.
Even so, the exposure that separated the groups is worth naming plainly, because it is unglamorous and reachable: roughly three structured, moderately hard hours a week, held for longer than a year. Not a protocol handed down by a paper — a description of what one group happened to have done. Our guide to building a longevity protocol is a reasonable place to organise it.
And keep the instruments in their place. A step count tells you that you moved; it does not tell you that anything was asked of you. That gap is where this study lives, and it is the wider point behind what a wearable can and cannot measure. Use numbers to correct fantasy, not to replace experience.
