Autophagy: What Can Actually Be Activated, and Measured

What Autophagy Actually Is

Autophagy is the process by which cells degrade and recycle their own components. Damaged organelles, misfolded proteins and aggregates are enclosed in a double-membrane vesicle, delivered to the lysosome and broken down, with the constituents reused.

Several forms exist: macroautophagy, the bulk process usually meant by the term; chaperone-mediated autophagy, which targets specific proteins; and selective forms including mitophagy for mitochondria and lipophagy for lipid droplets.

It is constitutive and continuous rather than a switch. Basal autophagy runs at all times as part of normal cellular housekeeping, and it is upregulated by nutrient scarcity, energy stress and the accumulation of damaged material.

Its decline is one of the twelve hallmarks of ageing, and impaired autophagy contributes to the accumulation of damaged mitochondria and protein aggregates. That much is well established, and it is why the process attracts so much interest.

The Signalling, and What Regulates It

mTORC1 is the principal negative regulator. When nutrients are abundant, particularly amino acids and especially leucine, mTORC1 is active, phosphorylates ULK1 and suppresses autophagy initiation.

AMPK is the principal positive regulator. Low cellular energy charge activates AMPK, which inhibits mTORC1 and directly activates ULK1.

Insulin signalling feeds into mTORC1 activation, which is why insulin rather than calories as such is one of the relevant inputs.

Amino acid availability is the most specific input. Leucine sensing through the mTORC1 pathway means protein intake suppresses autophagy more specifically than total energy does. This is the most practically important fact in the article, since it means a fast including protein is not producing the state the protocol intends.

Sirtuins and NAD+ availability also modulate it.

The mTOR article covers the pathway in more depth, and the AMPK article covers the other arm.

The Measurement Problem

The interpretive difficulty is worth understanding because it explains why no consumer test exists. Autophagy markers reflect the number of autophagosomes present, which rises either because more are being made or because fewer are being cleared, and those mean opposite things. Distinguishing them requires blocking the pathway and comparing, which cannot be done in a person.

Ketone measurement is the closest available proxy and it measures something else: a shift toward fat oxidation, which accompanies fasting and does not quantify autophagic activity. Anyone using a ketone meter to confirm autophagy is confirming a fasted state.

What Plausibly Increases It

Ranked by strength of the human case rather than by how often it is claimed.

Exercise. The best human evidence in this list. Muscle biopsy studies show autophagy markers change with exercise, and exercise-induced mitophagy is reasonably supported. It is also free, and it is rarely the intervention people reach for.

Energy restriction and fasting. Mechanistically sound through AMPK and mTORC1, demonstrated extensively in animals, and unquantified in humans. Longer probably means more, with no specifiable threshold.

Protein restriction, specifically leucine restriction. The most specific lever on mTORC1, and it conflicts directly with the muscle-preservation priority that matters in later decades. This is a genuine trade-off rather than a resolved one.

Sleep. Glymphatic clearance in the brain is most active during slow-wave sleep, and while glymphatic flow is not autophagy, sleep restriction impairs autophagy markers in animal studies.

Rapamycin. The most direct pharmacological mTORC1 inhibitor, and a prescription immunosuppressant rather than a supplement.

Spermidine. Induces autophagy in models, and observational data associate higher dietary intake with lower mortality. Human intervention data are limited and it is among the more interesting compounds here.

What has little: most compounds marketed as autophagy activators, including resveratrol and various polyphenols, where cell-culture activation occurs at concentrations oral dosing does not reach.

The Timing Conflict That Matters Most

Autophagy and muscle protein synthesis are driven by opposite signalling states, which creates a genuine scheduling problem rather than a theoretical one.

mTORC1 activation is required for muscle protein synthesis and suppresses autophagy. AMPK activation promotes autophagy and inhibits mTORC1. Pursuing both maximally at the same time is not possible.

The practical resolution is oscillation rather than optimisation. Feed protein and allow mTORC1 its window after resistance training, and place the scarcity signals, fasting, extended aerobic work, energy restriction, elsewhere in the day or week.

The stakes differ by age. In a 35-year-old the trade-off is roughly even. In a 75-year-old, sarcopenia risk means muscle preservation should generally win, and a protocol maximising autophagy through protein restriction is optimising the wrong pathway. This is the same age-dependence that runs through the IGF-1 question.

That framing produces a more defensible protocol than any hour count: periodic scarcity signalling, protein and loading protected, and no attempt to hold both states at once.

A Defensible Position

Given that the process is real, important and unmeasurable, the reasonable approach is to do the things with independent value that plausibly increase it, and to ignore the numbers.

Exercise regularly, both aerobic and resistance. Best human evidence, and worth doing for a dozen other reasons.

A 12-hour overnight fast aligned to sleep, which is nearly costless and provides periodic scarcity signalling.

Avoid chronic energy surplus, which keeps mTORC1 chronically active.

Protect sleep.

Occasional longer fasts if they suit you, with the safety caveats that apply beyond 24 hours, and without believing a specific hour count achieved anything specific.

Do not restrict protein to increase autophagy in later life, where the musculoskeletal cost is better established than the benefit.

Do not buy autophagy activator supplements, a test claiming to measure autophagy, or a protocol sold on hitting a threshold.

The honest summary: everything defensible in an autophagy protocol is something you would do anyway, and the parts that are specific to autophagy are the parts that cannot be verified. That is worth knowing before paying for the specificity.