Of the five autophagy-inducing practices, exercise is the one with the strongest human evidence and the most consistent dose-response. Fasting needs hours. Cold and sauna need protocols. Exercise is the lever you already pull every day, and the cellular cleanup it triggers is almost a byproduct of training you would do anyway.

This is the practitioner’s read on what exercise actually does at the cellular level, which modalities matter most, and how to dose it as a 40-plus athlete who wants to keep training hard for another 30 years.

What the human data actually says

Most of the autophagy literature on exercise comes from skeletal muscle biopsies in trained humans, taken before and after a session. That gives us cleaner evidence than the rodent work that drives most fasting claims. We can measure the rise in LC3-II, p62 turnover and mitophagy markers in real human tissue, and we can do it under controlled training conditions.

What the human evidence supports, with reasonable confidence:

• A single session of resistance training or high-intensity endurance work measurably increases autophagy markers in skeletal muscle, with the signal visible within 30 to 60 minutes post-exercise and persisting for several hours.
• Endurance work, particularly anything that taps into AMPK by depleting muscle glycogen, drives mitophagy, which is the selective recycling of damaged mitochondria. This is the mechanism behind the well-documented mitochondrial biogenesis effect of endurance training.
• Resistance training drives a broader autophagy response that includes the breakdown of damaged contractile proteins. Combined with the mTOR pulse from protein feeding afterwards, this is the cycle that grows healthier muscle, not just bigger muscle.
• Intensity matters more than duration for the acute autophagy signal. A 25-minute hard CrossFit metcon does more in this domain than a 90-minute moderate cycle.
• Fasted training amplifies the signal. Train with low glycogen and the AMPK response is bigger, the autophagy markers rise higher.

What the human evidence does not support, despite frequent claims:

• That you need a specific “longevity zone” of training to trigger autophagy. The signal turns up across resistance, HIIT, zone 2 and zone 4 work. Different flavours of stimulus, all of them effective.
• That more is better. Past a certain weekly training load, the autophagy response actually blunts as overtraining sets in. Recovery is when the cleanup happens; you cannot skip it.
• That zone 2 is uniquely magical for mitochondrial health. It is excellent and well-evidenced, but the marketing that has built up around it overstates how much it does that other endurance modalities do not.

Confidence in the above: high for the basic biology, high for the existence of an acute response, moderate for the comparative effects of different modalities, low to moderate for any claim about a specific weekly volume producing a specific long-term health outcome in an over-40 athletic population.

The cellular mechanism in one paragraph

Exercise lowers cellular energy charge, which activates AMPK. AMPK suppresses mTOR and switches on the autophagy machinery. Mechanical loading of muscle fibres creates micro-damage that is itself a trigger for selective autophagy of damaged proteins. The free radical burst from hard training, which sounds bad, is actually a signal that drives the antioxidant and autophagy response. Each of these inputs converges on the same cleanup pathway. Stack them together in a single session and the signal is larger than any single stressor on its own.

This is why CrossFit, which combines resistance loading, glycogen depletion and high lactate, produces such a strong acute response. It is also why a 40-minute mixed-modality session beats two separate hours of pure lifting and pure cardio for cellular cleanup, even if each isolated modality has its own merits.

How much, how hard: the practitioner’s protocols

Four patterns hold up across the literature and across actual 40-plus athletes who are still training well.

Three to four hard sessions a week, two to three easier ones

The default. Hard means metcon, threshold work, heavy lifting, hill repeats. Easy means zone 2, mobility, walking. The hard sessions drive the autophagy signal. The easy ones support recovery without adding to the load.

For most over-40 athletes, this works out to roughly six to eight hours of training a week, of which two to three hours is genuinely hard.

Zone 2 base plus occasional intensity

For those whose primary modality is endurance: four to six zone 2 sessions a week, 45 to 90 minutes each, plus one or two harder threshold or VO2 max sessions. The zone 2 work drives chronic mitochondrial adaptation. The hard sessions drive the acute autophagy spike.

Better suited to runners, cyclists, rowers, swimmers, anyone whose sport rewards aerobic depth.

Strength priority with hard finishers

For those who want to defend lean mass aggressively: three to four strength sessions a week as the main course, with a 10 to 15 minute hard conditioning piece tacked onto the end of each one. The strength work drives the mTOR/protein synthesis cycle. The finisher drives the AMPK/autophagy response. Compact and time-efficient.

This is what most over-40 strength-trained athletes I know who care about longevity have drifted towards.

Fasted morning training, fed sessions later

Fast through the night, train hard at 7:00 or 8:00, eat at noon. Once or twice a week, not every day. The combination of overnight fast and hard session creates the biggest combined autophagy signal you can produce without doing anything exotic. Cost: your performance on those sessions will be slightly below your fed best, so do not stack it onto your hardest weekly workout.

What does not work as an exercise-for-autophagy strategy: a 60-minute easy walk daily with no harder stimulus, “active recovery” yoga as your main weekly load, or seven hard sessions a week with no recovery days. The first is below threshold. The last produces overtraining, which suppresses the response you were trying to drive.

The athlete-specific concerns

Three real ones, in order of importance.

Recovery is where the cleanup happens. You can trigger autophagy in the gym, but the actual cellular maintenance work continues for 12 to 24 hours afterwards, peaking during sleep. Cut your sleep short and you cut the response short. An over-40 athlete trying to optimise this is better served by adding an hour of sleep than by adding a sixth weekly training session.

Soreness is not the metric. Excessive soreness signals damage that the autophagy machinery has to spend its capacity repairing, not optimising. The CrossFit reflex of grading a workout by how wrecked you feel afterwards works against you at 45 in a way it did not at 25. Quality of session beats brutality of session.

Protein still has to land properly. Hard training increases protein needs, and over-40 athletes are already at the higher end of the recommended range. If you train hard and then under-eat protein, you spend the autophagy response degrading muscle instead of upgrading it. 1.6 to 2.2 g/kg/day, split across two or three meals with at least 30 g each, holds across the literature for this population.

What I actually do

A typical week: four CrossFit sessions of 45 to 60 minutes, mixed metcon and strength, two of them genuinely hard and two moderate. One zone 2 session, usually a long walk or easy rower, 60 to 90 minutes. One day fully off. One or two of those CrossFit sessions are fasted, usually a Wednesday morning. I rotate which day is the hardest based on how the previous one felt and what HRV looks like.

No fancy supplements aimed at exercise-induced autophagy. The protocol is the stimulus. Sleep, protein and not training a sixth day are the inputs that actually move the needle.

The honest caveat

The cellular biology of exercise-induced autophagy is solid science. The translation from “autophagy markers go up after a hard session” to “you will live longer if you do this protocol” is several inferential steps and we do not have the 30-year trial to close the gap. The case for hard training as a longevity intervention rests on cellular, mechanistic and observational evidence, not on randomised long-term outcome data.

For an over-40 athlete the practical question is not whether exercise drives autophagy. It does. The question is whether you can sustain the training load that drives it without breaking yourself. The answer is yes, but only if recovery, sleep and nutrition are treated as part of the protocol, not as afterthoughts.

The cellular cleanup is real. The dose-response is well-evidenced. The version that works long-term is the boring sustainable one, not the seven-days-a-week hero training your Instagram feed told you about.

The app

An iPhone app of the same name is in development. It will track exercise alongside the other four autophagy practices, tag sessions by modality and intensity, and surface the cumulative weekly training load against your recovery markers from HealthKit. Launch will be announced here.